From f164ff588f222ff9a238d1b4da8ee9bb1844101b Mon Sep 17 00:00:00 2001 From: Mattk70 Date: Sun, 24 Nov 2024 15:56:36 +0000 Subject: [PATCH] initial settings translations, for many, chatGPT got lazy. Added setting for when the file timestamp is the start of the recording (h1E peculiarity) update to spanish context menu Exportar Audio(s) --- Help/{ebird.html => ebird.en.html} | 0 Help/ebird.es.html | 70 +++ Help/settings.da.html | 317 +++++++++++++ Help/settings.de.html | 320 +++++++++++++ Help/{settings.html => settings.en.html} | 0 Help/settings.es.html | 318 +++++++++++++ Help/settings.fr.html | 315 ++++++++++++ Help/settings.nl.html | 318 +++++++++++++ Help/settings.pt.html | 318 +++++++++++++ Help/usage.es.html | 4 +- I18n/community.de.json | 12 + I18n/i18n.zip | Bin 63946 -> 0 bytes I18n/index.da.json | 3 + I18n/index.de.json | 2 + I18n/index.en.json | 2 + I18n/index.es.json | 2 + I18n/index.fr.json | 2 + I18n/index.nl.json | 3 + I18n/index.pt.json | 2 + I18n/index.ru.json | 2 + I18n/index.sv.json | 2 + I18n/index.zh.json | 2 + I18n/settings.en.json | 578 ----------------------- index.html | 10 + js/i18n.js | 2 +- js/state.js | 3 +- js/ui.js | 27 +- 27 files changed, 2045 insertions(+), 589 deletions(-) rename Help/{ebird.html => ebird.en.html} (100%) create mode 100644 Help/ebird.es.html create mode 100644 Help/settings.da.html create mode 100644 Help/settings.de.html rename Help/{settings.html => settings.en.html} (100%) create mode 100644 Help/settings.es.html create mode 100644 Help/settings.fr.html create mode 100644 Help/settings.nl.html create mode 100644 Help/settings.pt.html create mode 100644 I18n/community.de.json delete mode 100644 I18n/i18n.zip delete mode 100644 I18n/settings.en.json diff --git a/Help/ebird.html b/Help/ebird.en.html similarity index 100% rename from Help/ebird.html rename to Help/ebird.en.html diff --git a/Help/ebird.es.html b/Help/ebird.es.html new file mode 100644 index 0000000..bef212a --- /dev/null +++ b/Help/ebird.es.html @@ -0,0 +1,70 @@ + + + + + Preguntas frecuentes sobre los registros de eBird + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Preguntas frecuentes sobre los registros de eBird
¿Qué es eBird?eBird es una plataforma online diseñada para compartir datos de observaciones de aves gestionada por el Departamento de Ornitología de la Universidad de Cornell (EE.UU.).
¿Qué es un archivo de registro de eBird?El archivo de registro es una hoja de cálculo con datos de observaciones de aves que se puede subir a la plataforma eBird. +
¿Cómo organiza Chirpity mis detecciones? +

La aplicación crea una lista de verificación para cada archivo asociado con la exportación. Estas listas contienen un total de detecciones de reclamos para cada especie del archivo. Si usted ha indicado un número concreto de reclamos, se utiliza esa cifra; de lo contrario, se asumirá que hay un solo reclamo.

+

El archivo exportado contiene una fila de cabecera que explica lo que representa cada columna. Usted puede consultar más detalles + en la web de eBird. +

+

Importante: Hay que eliminar la fila de cabecera antes de subir el archivo.

+
¿Cómo puedo subir mi archivo de registros de eBird? + Vaya a eBird y cree una cuenta o entre en su cuenta de eBird. Una vez tenga una cuenta y haya entrado, vaya al formulario de importación. +
    +
  1. Haga click en "Seleccionar archivo" y seleccione el archivo que haya preparado
    2. +
  2. Importante: elija "Formato (Extendido)"
    3. +
  3. Haga click en "Cargar archivo"
    4. +
+
¿Por qué da error cuando trato de subir el archivo? +

Primero, verifique que se haya eliminado la fila de cabecera; eBird sólo admite archivos en los que en la primera fila aparezca una observación.

+

Si ese no fuese el problema y usted detecta errores relacionados con especies no reconocidas, asegúrese de que las preferencias de idioma configuradas en Chirpity coincidan con las de su cuenta de eBird.

+
¿Por qué aparecen las columnas de Estado/Provincia en blanco? +

Estos campos son opcionales en el formato de registro de eBird. Aunque la API de búsqueda de lugares de OpenStreetMap utilizada por Chirpity proporciona el Estado y la provincia, no los ofrece en el formato utilizado por eBird. Usted puede buscar el formato correcto para su país o provincia en la siguiente hoja de cálculo.

+
¿Por qué has elegido eBird? ¿Tienes pensado que se puedan compartir registros con otras plataformas? +

eBird es la plataforma más sencilla de usar. Esto se debe a que tiene un formato de registro publicado y, quizá lo más importante, la taxonomía de eBird concuerda con la utilizada por los modelos BirdNET y Nocmig.

+

Espero poder ofrecer la posibilidad de compartir clips de audio con Xeno-Canto en futuras versiones. Esto dependerá de que ellos publiquen una API de carga adecuada.

+
+
La responsabilidad de verificar la exactitud del archivo de registro depende exclusivamente de usted. Chirpity no puede garantizar de ningún modo la precisión de los datos en los archivos de registro. Al usar esta función, usted acepta estos términos y está de acuerdo con ello.
+ + + \ No newline at end of file diff --git a/Help/settings.da.html b/Help/settings.da.html new file mode 100644 index 0000000..4828dad --- /dev/null +++ b/Help/settings.da.html @@ -0,0 +1,317 @@ + + + + + Chirpity Nocmig Indstillinger + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Detektioner
ModelVælg den model, der skal bruges til detektion. +
    +
  • + Nocmig bruger en model tilpasset natlige flyvekald fra europæiske fugle til analyse. +
    • +
  • + BirdNET bruger BirdNET-modellen udviklet af Stefan Kahl m.fl. fra K. Lisa Yang Center for Conservation Bioacoustics. +
    • +
+
TillidsgrænseDenne indstilling bestemmer det minimale tillidsniveau, der kræves for, at modellen rapporterer en detektion. + At øge værdien reducerer falske positiver, men øger risikoen for at gå glip af ægte kald. + NB. Tilliden kan justeres når som helst. Resultaterne opdateres selv efter analysen. +
Liste +

Lokale fugle. Udeluk fugle, der sandsynligvis ikke findes i dit område.

+

+ Når denne mulighed er valgt, vises et ekstra panel. Grænseværdi-indstillingen repræsenterer eBird-tjeklistefrekvensen. + For eksempel betyder en værdi på 0,03, at fugle, der forventes i mindre end 3 % af eBird-tjeklisterne i området, udelukkes. + Hvis du også markerer "Brug uge", udelukkes fugle, der ikke forventes på tidspunktet for optagelsen - dette læses fra filens tidsstempel. +
NB. I øjeblikket har de ugespecifikke filtre en begrænsning, hvor opsummeringen af detektioner bruger + listen fra den seneste fil i en batch af filer. + For eksempel, hvis du analyserer filer fra 22. april, 23. oktober og 24. februar i en batch, vil listen for 24. februar blive brugt til opsummeringen. +

+

Nataktive fugle. Dette udelukker detektioner af fuglesang og kald fra arter, der normalt ikke kalder om natten. + Når du aktiverer "Brug lokalitetsindstillinger" for BirdNET-modellen, vil listen omfatte fuglene på listen Lokale fugle, ekskluderet dem, + der ikke kalder om natten. +

+

Fugle. Inkluder detektioner for alle fugle, undtagen ikke-fugle-klasser.

+

Alt. Ud over alle fuglearter vises ikke-fugle detektioner, inklusive segmenter, der blev anset for støj.

+

Brugerdefineret. Lever en liste til modellen at bruge. Brugerdefinerede lister giver dig mulighed for:

+
    +
  1. Præcist at definere, hvilke detektioner du ønsker, at Chirpity skal rapportere.
    2. +
  2. Søge efter en bestemt art.
    3. +
+ Du kan bruge en forskellig brugerdefineret liste til både BirdNET og Nocmig-modeller. Formatet for en + brugerdefineret listefil er "videnskabeligt navn_dansk navn", med hver art på en ny linje. Vigtigt, brugerdefinerede lister skal bruge etiketter på det valgte sprog. + Du kan eksportere en startliste i det korrekte format fra Hjælp-menuen ved hjælp af linket "Hvilke arter detekteres?". +
Analyser kun natperioder + Søg kun efter detektioner om natten. Chirpity beregner begyndelsen og slutningen af den borgerlige skumring ud fra tidsstemplerne i dine lydfiler ved hjælp af den indstillede placering. + Hvis din optagelse spænder over både dag- og natperioder, springes dagperioderne over. Hvis du kun er interesseret i natkald, har dette to fordele: +
    +
  1. Der vil være færre falske positiver, da begge modeller laver flere fejl, når mange arter kalder samtidig (f.eks. under morgenkoret).
    2. +
  2. Analysen bliver hurtigere, da den ignorerer eventuelle dagperioder.
    3. +
+
Vis arter på IUCN's Rødliste i opsummeringen + Vis den globale rødliste-status for arterne i opsummeringstabellen, som vurderet af IUCN. Der er 7 kategorier: +
    +
  1. Mindst bekymring (LC)
    2. +
  2. Næsten truet (NT)
    3. +
  3. Sårbar (VU)
    4. +
  4. Truet (EN)
    5. +
  5. Kritisk truet (CR)
    6. +
  6. Uddød i naturen
    7. +
  7. Uddød
    8. +
+ Hvis du klikker på ikonet, åbnes vurderingssiden med arternes udbredelse og vurderingsdetaljer i din webbrowser. Dataene er hentet fra: + + IUCN 2024. IUCN's Rødliste over truede arter. Version 2024-2 + +
Ryd cache for referencekald + Metadata for arters sammenligninger, der bestemmer, hvad du ser, når du vælger "Sammenlign med referencekald" i en kontekstmenu, gemmes på disken. + Hvis du rydder cachen, slettes de gemte data, og artsdata vil blive genanmodet fra Xeno-Canto-webstedet, næste gang du beder om en sammenligning. + Dette kan være nyttigt, hvis en referencefil fjernes fra siden, eller hvis du ønsker at opdatere valget af kald. +
Standardplacering +
+ Kontekstmenuen, der kan tilgås fra filnavnet, indeholder muligheder for at ændre filens starttidspunkt og optagelsesplacering +
Kontekstmenu for filnavn.
+
+ Angiv standardplaceringen, der skal bruges til at beregne begyndelsen og slutningen af den borgerlige skumring samt generere din Lokale fugle-liste. + Denne indstilling kan tilsidesættes for individuelle filer ved at tildele en brugerdefineret placering i filnavnets kontekstmenu. +
System
sprog + Vælg sprog for "almindelige navne" etiketter, hvor oversættelser findes. De oversatte etiketter er hentet fra eBird. +
BackendDenne indstilling er kun synlig, når du bruger Nocmig-modellen. Vælg, hvordan forudsigelser beregnes på din computer. +
    +
  • + CPU bruger CPU'en til forudsigelser, hvilket generelt er hurtigere, hvis du har et integreret grafikkort. +
    • +
  • + WebGL bruger grafikkortet. Hvis du har et dedikeret grafikkort, vil dette typisk være + betydeligt hurtigere. +
    • +
  • + WebGPU Brug af en ny teknologi, der også bruger grafiske processorer. Det ligner WebGL, + men det er kendt for at fremskynde behandlingstiden på Mac-computere, i modsætning til WebGL. WebGPU bruger mere hukommelse end WebGL, og + hvis tråde eller batchstørrelsesværdierne er sat for højt, kan Chirpity løbe tør for hukommelse. Hvis dette sker, + kan spektrogrammet forsvinde fra skærmen, eller appen kan blive ufølsom. Brug en lavere værdi for tråde og / eller batchstørrelse. +
    • +
+
TrådeDenne indstilling giver dig mulighed for at vælge antallet af tråde, der skal bruges til forudsigelser. + Når CPU-backend'en bruges, sættes antallet af tråde automatisk til at matche antallet af CPU-kerner, der er tilgængelige på computeren. + Som standard bruger GPU-backend'en to tråde. Justering af denne indstilling sammen med batchstørrelsen kan øge behandlingshastigheden. +
BatchstørrelseDenne indstilling bestemmer antallet af tre-sekunders lydklip, der behandles for hver modelanmodning. + Større værdier kan føre til hurtigere behandling, især på lange filer, selvom for store værdier kan reducere behandlingshastigheden og endda forårsage hukommelsesproblemer. + Den optimale værdi kan også variere afhængigt af filformatet, der analyseres. +
Aktiver meddelelser om afsluttet analyse + Dette er nyttigt, hvis du vil køre Chirpity-analyse i baggrunden, + eller med vinduet minimeret. En systemadvarsel vil underrette dig, når analysen er afsluttet. +
Audioindstillinger
Justering af volumenHvis dine optagelser er meget stille, kan du øge lydstyrken ved at hæve gain. Lydstyrken kan øges med op til 50 decibel.
Normaliser lyd + Hvis aktiveret, vil niveauerne i lydfilen blive normaliseret. Kort sagt vil dette maksimere volumenområdet mellem de stilleste og højeste lyde. + Effekten vil sikre, at al lyd afspilles med en ensartet volumen, og vokaliseringer i stille lydfiler vil være lettere at høre. +
High Pass filterDette filter fjerner lyde under den valgte frekvens. + Primært beregnet til optagelser lavet i urbane omgivelser, undertrykker det lavfrekvent summen. +
Low Shelf filterDette filter ligner High Pass-filteret. Dog reducerer det ikke kun støjen, men sænker volumen af lavfrekvent støj. + Fordelen ved et low shelf filter er, at du kan reducere summen, samtidig med at du stadig kan høre arter, der har lavfrekvente kald. + Du kan justere tærskelfrekvensen og styrken af filteret (dæmpning). +
Audio Export
Format og BitrateVælg mellem lossless eller lossy formater for de lydklip, du eksporterer fra Chirpity.
Konverter til monoKombiner stereo spor til én kanal. Dette vil resultere i mindre filstørrelser.
DekoratorerDisse to muligheder kan bruges til at: +
    +
  • + Pad: Tilføj to sekunder til begyndelsen og slutningen af den valgte lyd til eksport +
    • +
  • + Fade: Anvend en fade-effekt på begyndelsen og slutningen af dine eksporterede filer. +
    • +
+
Audio Bibliotek
Tilføj til bibliotek ved gemningNår du gemmer et sæt resultater til arkivdatabasen, vil de tilhørende lydfiler blive konverteret til bibliotekformat og gemt på arkivplaceringen.
Skær lyd i biblioteket til natperioderDette vil beskære begyndelsen eller slutningen af en fil for at fjerne den lyd, der blev optaget i dagslys. Specifikt før slutningen af den borgerlige skumring om aftenen og efter slutningen af den borgerlige skumring om morgenen. + Disse perioder er indstillet i henhold til filens lokalitetsindstillinger. +
Bibliotekets placeringVælg en mappe til at opbevare lydbiblioteket. Filene vil blive organiseret i henhold til følgende struktur: + 
+        <Den valgte placering> / Optagelsessted / År / Måned / filnavn
+

For eksempel: D:\Vögel\56075, Karthäuserhofgelände\2024\Juli\20240704_0130.flac

+
BibliotekformatBiblioteket kan gemme din lyd i et lossy eller lossless format. +
    +
  • + Lossy: Lyd konverteres til Ogg Vorbis. Dette er den anbefalede mulighed. Lyden vil være praktisk talt uadskillelig fra originalen i applikationen - både for modellen og for det menneskelige øre. + Filstørrelsen vil blive dramatisk reduceret - så meget, at 24 timers optagelser vil kræve omkring 1 gigabyte lagerplads. +
    • +
  • + Lossless: Hvis lossless-optionen vælges, vil lydens troværdighed (bitrate, kanaler og sample rate) være identisk med den originale fil - selv WAV-filer. Ved at bruge FLAC-formatet vil de arkiverede filer + være cirka halvdelen af størrelsen af deres WAV-ækvivalenter. +
    • +
+
Spektrogramindstillinger
FarvekortVælg et farvetema til spektrogrammet, eller lav dit eget. Hvis du vælger 'brugerdefineret', får du mulighed for at indstille farverne for peak-, midt- og stille lyde efter + personlig præference. Du kan også justere midtpunktets position: med en værdi på 0 eller 1 vil spektrogrammet være tofarvet. Værdier imellem vil blande de tre farver. + Hvis du sætter Mid-farven den samme som en af de andre, vil du kunne justere kontrasten i spektrogrammet ved at bruge Mid Position justeringerne. +

I kombination med justering af audiofiltre giver et brugerdefineret farvekort dig mulighed for at forbedre kontrasten / synligheden af kald ved at bruge de farver, du ønsker.

+
VinduesfunktionEn række vinduesfunktioner er tilgængelige for spektrogrammet. Hver af dem har lidt forskellige egenskaber, så ændring af vinduesfunktionen kan også forbedre + udseendet af kald i visningen. +
TidslinjeSpektrogrammets tidslinje har to tilstande: +
    +
  • + Tidskode viser den tid, der er gået fra begyndelsen af filen. +
    • +
  • + Tid på dagen bruger filens tidsstempler til at bestemme den faktiske tid for de detekterede kald. + Hvis en fils tidsstempel ikke afspejler optagelsestiden, vil læsningen være unøjagtig. + For at redigere starttiden for den aktuelle fil skal du indstille den ønskede dato og tid ved hjælp af filnavnets kontekstmenu. +
    • +
+
Avanceret
Send filtreret lyd til analyse + Som standard bruges filterindstillingerne i Audioindstillingerne kun til at ændre lyden i applikationens grænseflade. Hvis du aktiverer denne indstilling, + vil den ændrede lyd blive brugt til analyse. Dette er normalt en dårlig idé, da det kan forringe modellernes nøjagtighed i de fleste situationer. Men hvis dine + optagelser har en betydelig mængde lavfrekvent støj, kan det måske være, at ændring af lyden ved hjælp af disse filtre forbedrer modellernes evne til + at lave nøjagtige detektioner. +
Konteksttilstand Denne tilstand er kun tilgængelig for Nocmig-modellen. Når den aktiveres, vil modellen bruge den omgivende kontekst, når den laver forudsigelser. Dette hjælper med at mindske + falske positive detektioner. +
SNR filterDenne indstilling er også kun tilgængelig for Nocmig-modellen. SNR (Signal-to-Noise Ratio) filteret kan kun aktiveres, når du bruger CPU-backend'en. + Når det er aktiveret, vil Chirpity ignorere lydsegmenter uden et klart lydhændelse. + Et stærkere signal kræves for, at en forudsigelse forsøges, når højere signal-til-støje-værdier er indstillet. + Formålet er at levere betydelige hastighedsforøgelser, men denne indstilling kan få Chirpity til at gå glip af stille eller fjerne kald. +
Debug-tilstand + Hvis du har stødt på et problem, mens du bruger Chirpity, kan du aktivere debug-tilstand. Når du ændrer denne indstilling, skal du afslutte og genstarte applikationen + for at ændringen kan træde i kraft. Når den startes i debug-tilstand, åbnes to vinduer, og hver af dem vil printe debug-information til en konsol på højre side af vinduet. + Dette er meget nyttigt, hvis du ønsker at rapportere en fejl - da du kan dele et skærmbillede af eventuelle fejl, der vises (disse vil blive vist i rød i konsollen). +
+ + \ No newline at end of file diff --git a/Help/settings.de.html b/Help/settings.de.html new file mode 100644 index 0000000..b6ad7a8 --- /dev/null +++ b/Help/settings.de.html @@ -0,0 +1,320 @@ + + + + + Chirpity Nocmig Einstellungen + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Erkennungen
ModellWählen Sie das Modell aus, das für die Erkennung verwendet werden soll. +
    +
  • + Nocmig verwendet ein Modell, das für die nächtlichen Flugrufe europäischer Vögel zur Analyse abgestimmt ist. +
    • +
  • + BirdNET verwendet das BirdNET Modell, das von Stefan Kahl et al. vom K. Lisa Yang Center for Conservation Bioacoustics entwickelt wurde. +
    • +
+
VertrauensschwelleDiese Einstellung bestimmt das minimale Vertrauensniveau, das erforderlich ist, damit das Modell eine Erkennung meldet. + Eine Erhöhung des Werts reduziert falsch-positive Ergebnisse, erhöht jedoch das Risiko, echte Rufe zu verpassen. + Hinweis: Das Vertrauen kann jederzeit angepasst werden. Die Ergebnisse werden auch nach der Analyse aktualisiert. +
Liste +

Lokale Vögel. Schließt Vögel aus, die in Ihrer Region voraussichtlich nicht vorkommen.

+

+ Wenn diese Option ausgewählt ist, erscheint ein zusätzliches Panel. Der Schwellenwert repräsentiert die Häufigkeit der eBird-Checklisten. + Ein Wert von 0,03 bedeutet also, dass Vögel, die in weniger als 3 % der eBird-Checklisten in der Region vorkommen, ausgeschlossen werden. + Wenn Sie auch "Woche verwenden" aktivieren, werden Vögel ausgeschlossen, die zu der Jahreszeit, in der die Aufnahme gemacht wurde, nicht erwartet werden - dies wird vom Zeitstempel der Datei abgelesen. +
Hinweis: Derzeit hat die wöchentlich spezifische Filterung eine Einschränkung, dass die Zusammenfassung der Erkennungen die + Liste aus der neuesten Datei in einer Batch von Dateien verwendet. + Zum Beispiel, wenn Sie Dateien vom 22. April, 23. Oktober und 24. Februar in einer Batch analysieren, wird die Liste vom 24. Februar für die Zusammenfassung verwendet. +

+

Nachts aktive Vögel. Dies schließt Erkennungen von Vogelgesang und den Rufen von Arten aus, die typischerweise nicht in der Nacht vokalisieren. + Wenn Sie "Verwenden Sie Standorteinstellungen" für das BirdNET-Modell aktivieren, wird die Liste die Vögel der Lokalen Vögel-Liste umfassen, + mit Ausnahme derer, die nicht nachts rufen.

+

Vögel. Schließt Erkennungen für alle Vögel ein, mit Ausnahme nicht-avianer Klassen.

+

Alles. Zeigt zusätzlich zu allen Vogelarten auch Nicht-Vogel-Erkennungen an, einschließlich Segmente, die als Rauschen betrachtet wurden.

+

Benutzerdefiniert. Geben Sie eine Liste an, die das Modell verwenden soll. Benutzerdefinierte Listen ermöglichen es Ihnen,:

+
    +
  1. Die Erkennungen präzise zu definieren, die Sie von Chirpity gemeldet haben möchten
    2. +
  2. Nur nach einer bestimmten Art zu suchen
    3. +
+ Sie können eine andere benutzerdefinierte Liste für das BirdNET- und Nocmig-Modell verwenden. Das Format einer + benutzerdefinierten Listen-Datei ist "wissenschaftlicher Name_gängiger Name", mit jeder Art auf einer neuen Zeile. Wichtig ist, dass die benutzerdefinierte Liste die Bezeichnungen in Ihrer ausgewählten Sprache verwenden muss. + Sie können eine Startliste im richtigen Format aus dem Hilfemenü exportieren, indem Sie auf den Link "Welche Arten werden erkannt?" klicken. +
Nur Nachtzeiten analysieren + Suchen Sie nur nach Erkennungen während der Nacht. Chirpity berechnet den Beginn und das Ende der bürgerlichen Dämmerung aus den Zeitstempeln Ihrer Audiodateien unter Verwendung des von Ihnen festgelegten Standorts. + Wenn Ihre Aufnahme sowohl Tages- als auch Nachtzeiträume umfasst, werden die Tageszeiträume übersprungen. Wenn Sie nur an nächtlichen Rufen interessiert sind, hat dies zwei Vorteile: +
    +
  1. Es wird weniger falsch-positive Erkennungen geben, da beide Modelle mehr Fehler machen, wenn viele Arten gleichzeitig vokalisieren (z. B. während des Morgengesangs).
    2. +
  2. Die Analyse wird schneller abgeschlossen, da sie den Tageszeitraum ignoriert.
    3. +
+
Zeige IUCN Rote Liste Status der Arten in der Zusammenfassung + Zeigen Sie den globalen Status der Roten Liste der Arten in der Zusammenfassungstabelle an, wie er von der IUCN bewertet wurde. Es gibt 7 Kategorien: +
    +
  1. Geringe Besorgnis (LC)
    2. +
  2. Fast bedroht (NT)
    3. +
  3. Vulnerabel (VU)
    4. +
  4. Gefährdet (EN)
    5. +
  5. Kritisch gefährdet (CR)
    6. +
  6. In der Wildnis ausgestorben
    7. +
  7. Ausgestorben
    8. +
+ Wenn Sie auf das Symbol klicken, wird die Bewertungsseite mit dem Verbreitungsgebiet der Art und den Details der Bewertung in Ihrem Webbrowser geöffnet. Die Daten stammen von: + + IUCN 2024. IUCN Rote Liste der bedrohten Arten. Version 2024-2 + +
Cache für Referenzaufrufe löschen + Die Metadaten der Artenspezifikationen, die bestimmen, was Sie sehen, wenn Sie „Mit Referenzaufrufen vergleichen“ im Kontextmenü auswählen, werden auf der Festplatte gespeichert. Wenn Sie den Cache löschen, + werden die gespeicherten Daten gelöscht, und Artendaten werden bei der nächsten Anfrage zur Vergleichsanalyse von der Xeno-Canto-Website erneut abgerufen. + Dies kann nützlich sein, wenn eine Referenzdatei von der Website entfernt wurde oder wenn Sie die Auswahl der Rufe aktualisieren möchten. +
Standardstandort +
+ Das Kontextmenü, das über den Dateinamen aufgerufen wird, enthält Optionen zur Änderung der Startzeit der Datei und des Aufnahmeorts +
Kontextmenü für den Dateinamen.
+
+ Legen Sie den Standardstandort fest, der zur Berechnung des Beginns und Endes der bürgerlichen Dämmerung sowie zur Generierung Ihrer Lokalen Vögel-Liste verwendet werden soll. + Diese Einstellung kann für einzelne Dateien überschrieben werden, indem ein benutzerdefinierter Standort im Kontextmenü des Dateinamens zugewiesen wird. +
System
Sprache + Wählen Sie die Sprache für die "häufigen Namen"-Bezeichnungen, wo Übersetzungen vorhanden sind. Die übersetzten Bezeichnungen wurden von eBird bereitgestellt. +
BackendDiese Option ist nur sichtbar, wenn das Nocmig-Modell verwendet wird. Wählen Sie, wie Vorhersagen auf Ihrem Computer berechnet werden. +
    +
  • + CPU verwendet die CPU für Vorhersagen. Dies ist in der Regel schneller, wenn Sie eine integrierte + Grafikkarte haben. +
    • +
  • + WebGL verwendet die Grafikkarte. Wenn Sie eine dedizierte Grafikkarte haben, ist dies in der Regel + erheblich schneller. +
    • +
  • + WebGPU Verwendet eine neue Technologie und nutzt ebenfalls Grafikprozessoren. Es ähnelt WebGL, + aber es ist bekannt dafür, die Verarbeitung auf Macs zu beschleunigen, im Gegensatz zu WebGL. WebGPU verwendet mehr Speicher als WebGL, und + wenn die Threads oder die Batchgröße zu hoch eingestellt sind, kann Chirpity aus dem Speicher laufen. Wenn dies geschieht, + kann das Spektrogramm vom Bildschirm verschwinden oder die App kann nicht mehr reagieren. Verwenden Sie einen niedrigeren Wert für Threads und / oder Batchgröße. +
    • +
+
ThreadsDiese Einstellung ermöglicht es Ihnen, die Anzahl der Threads auszuwählen, die für Vorhersagen verwendet werden sollen. + Bei Verwendung des CPU-Backends wird die Anzahl der Threads automatisch auf die Anzahl der verfügbaren CPU-Kerne auf dem Computer eingestellt. + Standardmäßig verwendet das GPU-Backend zwei Threads. Durch Anpassen dieser Einstellung zusammen mit der Batchgröße kann die Verarbeitungsgeschwindigkeit erhöht werden. +
BatchgrößeDiese Einstellung bestimmt die Anzahl der drei Sekunden langen Audiostücke, die für jede Modellanfrage verarbeitet werden. + Größere Werte können zu einer schnelleren Verarbeitung führen, besonders bei langen Dateien, obwohl zu große Werte die Verarbeitungsgeschwindigkeit verringern und sogar zu Speicherproblemen führen können. + Der optimale Wert kann auch je nach dem analysierten Dateiformat variieren. +
Aktivieren Sie Benachrichtigungen über die Analyseabschlüsse + Dies ist nützlich, wenn Sie Chirpity-Analysen im Hintergrund ausführen möchten, + oder wenn das Fenster minimiert ist. Eine Systembenachrichtigung informiert Sie, wenn die Analyse abgeschlossen ist. +
Audioeinstellungen
LautstärkeanpassungWenn Ihre Aufnahmen sehr leise sind, können Sie die Lautstärke erhöhen, indem Sie die Verstärkung erhöhen. Die Lautstärke kann um bis zu 50 Dezibel erhöht werden.
LautstärkeanpassungWenn Ihre Aufnahmen sehr leise sind, können Sie die Lautstärke erhöhen, indem Sie die Verstärkung erhöhen. Die Lautstärke kann um bis zu 50 Dezibel erhöht werden.
Audio normalisieren + Wenn aktiviert, werden die Pegel in der Audiodatei normalisiert. Einfach ausgedrückt maximiert dies den Lautstärkebereich zwischen den leisesten und lautesten Tönen. + Die Wirkung stellt sicher, dass alle Audiodateien mit ähnlicher Lautstärke abgespielt werden und Vokalisierungen in leisen Audiodateien leichter hörbar sind. +
High Pass FilterDieses Filter entfernt Geräusche unterhalb der ausgewählten Frequenz. + Es ist hauptsächlich für Aufnahmen aus städtischen Gebieten gedacht und unterdrückt niederfrequente Brummgeräusche. +
Low Shelf FilterDieses Filter ist dem High Pass Filter ähnlich. Es reduziert jedoch nicht nur Geräusche, sondern senkt auch die Lautstärke von niederfrequenten Geräuschen. + Der Vorteil eines Low Shelf Filters ist, dass Sie das Brummen verringern können, während Sie weiterhin Arten hören können, die niederfrequente Rufe haben. + Sie können die Schwellenfrequenz und die Stärke des Filters (Dämpfung) einstellen. +
Audio Export
Format und BitrateWählen Sie aus verlustfreien oder verlustbehafteten Formaten für die Audioausschnitte, die Sie aus Chirpity exportieren.
In Mono konvertierenKombinieren Sie Stereo-Tracks zu einem einzelnen Kanal. Dies führt zu kleineren Dateigrößen.
DekoratorenDiese beiden Optionen können verwendet werden, um: +
    +
  • + Padding: Fügen Sie zwei Sekunden am Anfang und Ende des für den Export ausgewählten Audios hinzu. +
    • +
  • + Fade: Wenden Sie einen Fade-Effekt auf den Anfang und das Ende Ihrer exportierten Dateien an. +
    • +
+
Audio-Bibliothek
Beim Speichern zur Bibliothek hinzufügenWenn Sie eine Ergebnismenge in der Archivdatenbank speichern, werden die zugehörigen Audiodateien in das Bibliotheksformat konvertiert und am Archivstandort gespeichert.
Bibliotheks-Audio auf Nachtzeiten zuschneidenDies schneidet den Anfang oder das Ende einer Datei ab, um das Audio zu entfernen, das während des Tages aufgenommen wurde. Insbesondere vor dem Ende der bürgerlichen Dämmerung am Abend und nach dem Ende der bürgerlichen Dämmerung am Morgen. + Diese Zeiträume werden gemäß den Standorteinstellungen der Datei festgelegt.
BibliotheksstandortWählen Sie einen Ordner aus, in dem Sie die Audio-Bibliothek speichern möchten. Die Dateien werden gemäß der folgenden Struktur organisiert: + 
+        <Ihr gewählter Standort> / Aufnahmeort / Jahr / Monat / Dateiname
+

Zum Beispiel: D:\Vögel\56075, Karthäuserhofgelände\2024\Juli\20240704_0130.flac

+
BibliotheksformatDie Bibliothek kann Ihre Audio-Dateien in einem verlustbehafteten oder verlustfreien Format speichern. +
    +
  • + Verlustbehaftet: Audio wird in Ogg Vorbis konvertiert. Dies ist die empfohlene Option. Das Audio wird in der Anwendung praktisch nicht vom Original zu unterscheiden sein – sowohl für das Modell als auch für das menschliche Ohr. + Die Dateigröße wird dramatisch reduziert – so dass 24 Stunden Aufnahmen etwa 1 GB Speicherplatz benötigen. +
    • +
  • + Verlustfrei: Wenn die verlustfreie Option ausgewählt wird, bleibt die Audioqualität (Bitrate, Kanäle und Abtastrate) identisch mit der Originaldatei – sogar WAV-Dateien. Durch die Verwendung des FLAC-Formats sind die archivierten Dateien + etwa halb so groß wie ihre WAV-Äquivalente. +
    • +
+
Spektrogrammeinstellungen
FarbenkarteWählen Sie ein Farbthema für die Spektrogrammanzeige oder erstellen Sie Ihr eigenes. Wenn Sie „Benutzerdefiniert“ auswählen, haben Sie die Möglichkeit, die Farben für Spitzen-, Mittel- und leise Töne nach + persönlichen Vorlieben einzustellen. Sie können auch die Position des Mittelpunkts anpassen: Mit einem Wert von 0 oder 1 wird das Spektrogramm zweifarbig. Werte dazwischen mischen die drei Farben. + Wenn Sie die Mittel-Farbe auf eine der anderen Farben setzen, können Sie den Kontrast im Spektrogramm mithilfe der Anpassungen der Mittel-Position einstellen. +

In Kombination mit der Anpassung von Audiofiltern ermöglicht eine benutzerdefinierte Farbenkarte, den Kontrast / die Sichtbarkeit von Rufen mit den Farben Ihrer Wahl zu verstärken.

+
FensterfunktionEine Reihe von Fensterfunktionen sind für die Spektrogrammanzeige verfügbar. Jede hat leicht unterschiedliche Eigenschaften, sodass das Ändern der Fensterfunktion auch das + Aussehen der Rufe in der Anzeige verbessern kann. +
ZeitachseDie Spektrogramm-Zeitachse hat zwei Modi: +
    +
  • + Zeitscode zeigt die seit Beginn der Datei vergangene Zeit an. +
    • +
  • + Tag und Uhrzeit verwendet die Zeitstempel der Datei, um die tatsächliche Zeit der erkannten Rufe zu bestimmen. + Wenn der Zeitstempel einer Datei nicht die Aufnahmezeit widerspiegelt, ist die Anzeige ungenau. + Um die Startzeit der aktuellen Datei zu bearbeiten, stellen Sie das gewünschte Datum und die Uhrzeit im Kontextmenü des Dateinamens ein. +
    • +
+
Erweitert
Gefilterte Audiodaten zur Analyse senden + Standardmäßig werden die Filtereinstellungen in den Audioeinstellungen nur verwendet, um die Audiodateien in der Anwendung zu ändern. Wenn Sie diese Einstellung aktivieren, + wird das bearbeitete Audio für die Analyse verwendet. Dies ist normalerweise keine gute Idee, da es die Genauigkeit der Modelle in den meisten Situationen beeinträchtigt. Wenn jedoch Ihre + Aufnahmen eine beträchtliche Menge an niederfrequentem Rauschen enthalten, kann das Bearbeiten des Audios mit diesen Filtern die Fähigkeit der Modelle verbessern, + genaue Erkennungen vorzunehmen. +
Kontextmodus Dieser Modus ist nur für das Nocmig-Modell verfügbar. Wenn aktiviert, verwendet das Modell die umgebende Kontextinformation, wenn es Vorhersagen trifft. Dies hilft, + falsche positive Erkennungen zu reduzieren. +
SNR-FilterDiese Einstellung ist ebenfalls nur für das Nocmig-Modell verfügbar. Der SNR (Signal-to-Noise Ratio)-Filter kann nur aktiviert werden, wenn das CPU-Backend verwendet wird. + Wenn aktiviert, ignoriert Chirpity Audiosegmente ohne ein deutliches Soundereignis. + Ein stärkeres Signal ist erforderlich, damit eine Vorhersage versucht wird, wenn höhere Signal-zu-Rausch-Werte eingestellt sind. + Der Zweck dieser Einstellung ist es, signifikante Geschwindigkeitsgewinne zu erzielen, jedoch kann diese Einstellung dazu führen, dass Chirpity leisere oder entferntere Rufe übersieht. +
Debug-Modus + Wenn Sie ein Problem bei der Verwendung von Chirpity festgestellt haben, können Sie den Debug-Modus aktivieren. Wenn Sie diese Einstellung ändern, müssen Sie die Anwendung schließen und neu starten, + damit die Änderung wirksam wird. Wenn die Anwendung im Debug-Modus gestartet wird, öffnen sich zwei Fenster, und jedes gibt Debugging-Informationen in einer Konsole auf der rechten Seite des Fensters aus. + Dies ist extrem nützlich, wenn Sie einen Fehler melden möchten – da Sie einen Screenshot von allen Fehlern teilen können, die erscheinen (diese werden in der Konsole rot angezeigt). +
+ + \ No newline at end of file diff --git a/Help/settings.html b/Help/settings.en.html similarity index 100% rename from Help/settings.html rename to Help/settings.en.html diff --git a/Help/settings.es.html b/Help/settings.es.html new file mode 100644 index 0000000..dc9b8ed --- /dev/null +++ b/Help/settings.es.html @@ -0,0 +1,318 @@ + + + + + Chirpity Nocmig Einstellungen + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Erkennungen
ModelWählen Sie das Modell für die Erkennung aus. +
    +
  • + Nocmig wird ein Modell verwenden, das auf die nächtlichen Flugrufe europäischer Vögel abgestimmt ist. +
    • +
  • + BirdNET wird das BirdNET Modell verwenden, das von Stefan Kahl et al. vom K. Lisa Yang Center for Conservation Bioacoustics entwickelt wurde +
    • +
+
VertrauensschwelleDiese Einstellung legt das minimale Vertrauensniveau fest, das das Modell für die Meldung einer Erkennung benötigt. + Ein höherer Wert reduziert falsch-positive Ergebnisse, erhöht jedoch das Risiko, echte Rufe zu übersehen. + N.B. Das Vertrauensniveau kann jederzeit angepasst werden. Die Ergebnisse werden auch nach der Analyse aktualisiert. +
Listae

Lokale Vögel. Exclude birds unlikely to be found in your location.

+

With this option selected, an additional panel appears. The Threshold value represents the eBird checklist frequency. + So, a value of 0.03 means birds that are predicted to occur in less than 3% of eBird checklists in the area are excluded. + If you also check "Use Week", it excludes birds not expected at the time of year the recording was made - which is read from the file's timestamp. +
N.B. Currently, a limitation to the week specific filters is that the summary of detections wird das + list from the latest file in a batch of files. + For example, if you analyse files from April 22, October 23 and February 24 in a batch, the list for February 24 will be used for the summary. +

+

Nocturnal Birds. This will exclude detections of bird song and the calls of species which do not typically vocalise during the night. + When you enable "Use Location settings" for the BirdNET model, the list will comprise the birds on the Lokale Vögel list, excluding those + which do not call at night.

+

Birds. Include detections for all birds, excluding non-avian classes. +

+

Everything. In addition to all bird species, show non-bird detections including segments that were considered noise.

+

Custom. Provide a list for the model to use. Custom lists allow you to:

+
    +
  1. Precisely define the detections you would like Chirpity to report on
    2. +
  2. Search only for a specific species
    3. +
+ You may use a different custom list for both BirdNET and Nocmig models. The format for a + custom list file is "scientific name_common name", with each species on a new line. Importantly, the custom list needs to use labels in your selected language. You can export a starting + list in the correct format from the Help menu, using the "What species are detected?" link. +
Only analyse night time periods + Only search for detections at night. Chirpity calculates the end and start of civil twilight from the timestamps of your audio files using the location you have set. If your recording + spans both daylight and night time periods, the day periods will be skipped. If you are only interested in nocturnal calls this has two benefits: +
    +
  1. There will be fewer false positives, since both models make more mistakes when many species vocalise at once (e.g. during the dawn chorus)
    2. +
  2. The analysis will complete faster, since it ignores any daytime period
    3. +
+
Show Species IUCN Red Listae status in the summary + Display the global Red Listae status of the species in the summary table, as assessed by the IUCN There are 7 categories: +
    +
  1. Least Concern (LC)
    2. +
  2. Near Threatened (NT)
    3. +
  3. Vulnerable (VU)
    4. +
  4. Endangered (EN)
    5. +
  5. Critically ENdangered (CR)
    6. +
  6. Extinct in the Wild
    7. +
  7. Extinct
    8. +
+ If you click the icon, it will open the assessment page with the species' range and details of the assessment in your web browser. The data is sourced from: + + IUCN 2024. IUCN Red Listae of Threatened Species. Version 2024-2 + +
Clear reference call cache + The species' comparison metadata that determine what you see when you select "Compare with Reference Calls" on a context menu are saved to disk. If you clear the cache, + the saved data are deleted, and species data will be re-requested from the Xeno-Canto website when you next ask for a comparison. This may be useful if a reference file + is removed from the site, or if you would like to update the selection of calls. +
Default Location +
+ The context menu accessed from the filename contains options to amend the file's start time and the recording location +
File name context menu.
+
+ Set the default location to use for calculating the start and end of civil twilight and also generating your Lokale Vögel list. This setting can be overridden for individual + files by assigning a custom location in the filename context menu. +
System
Language + Choose the language for the "common name" labels, where translations exist. The translated labels have been obtained from eBird. +
BackendThis option is only visible when using the Nocmig model. Choose how predictions are calculated on you computer. +
    +
  • + CPU wird das CPU for predictions, this is generally faster if you have an integrated + graphics card. +
    • +
  • + WebGL wird das graphics card. If you have a dedicated graphics card, this will typically be + significantly faster. +
    • +
  • + WebGPU This employs a new technology and will also use graphics processors. It is similar to WebGL, + but is known to speed up the processing times on Macs, unlike WebGL. WebGPU uses more memory than WebGL, and + if the threads or batch size values are set too high, Chirpity may run out of memory. If this happens + the spectrogram may disappear from the screen or the app may become unresponsive. Use a lower value for threads and / or batch size. +
    • +
+
ThreadsThis setting allows you to select the number of threads to use for predictions. + When using the CPU backend, the number of threads is automatically set to match the number of CPU cores available on the computer. + By default, the GPU backend uses two threads. Adjusting this setting, along with the batch size, can increase the speed of processing. +
Batch SizeThis setting determines the number of three-second audio chunks to process for each model request. + Larger values may lead to faster processing, especially on long files, although too large a value will reduce the processing speed and may even cause memory issues. + The optimal value may also differ depending on the file format being analysed. +
Enable analysis completion notifications + This is useful if you would like to run Chirpity analysis in the background, + or with the window minimised. A system alert will notify you when the analysis completes. +
Audio Preferences
Volume AdjustmentIf your recordings are very quiet, you can increase the loudness of the audio by increasing the gain. The volume of audio can be increased by up to 50 decibels.
Normalise audio + If enabled, the levels in the audio file will be normalised. In simple terms, this will maximise the volume range between the quietest and loudest sounds. + The effect will ensure all audio plays at a similar volume, and vocalisations in quiet audio files will be easier to hear. +
High Pass filter This filter removes the sound below the frequency selected. + Intended primarily for recordings made in in urban settings, this suppresses low-frequency hum. +
Low Shelf filter This filter is similar to the High Pass filter. However, rather than simply remove noise, it reduces the volume of low frequency noise. + The advantage of a low shelf filter is that you can reduce the hum, whilst still being able to hear species which have low frequency calls. + You can set the threshold frequency, and the strength of the filter to apply (Attenuation). +
Audio Export
Format and BitrateChoose from lossless or lossy formats for the audio clips you export from Chirpity.
Convert to monoCombine stereo tracks to a single channel. This will result in smaller file sizes.
DecoratorsThese two options can be used to: +
    +
  • + Pad: Add two seconds to the beginning and end of the audio selected for export +
    • +
  • + Fade: Apply a fade effect to the beginning and end of your exported files. +
    • +
+
Audio Library
Add to Library on SaveWhen you save a set of results to the archive database, the associated audio files will be converted to the library format and saved in the archive location.
Crop library audio to night time periodsThis will crop the start or end sections of a file to remove the audio that was recorded during daylight. Specifically, before the end of Civil Twilight in the evening and after the end of Civil Twilight in the morning. + Those periods are set according to the file's location settings.
Library LocationSelect a folder in which to store the audio library. The files will be organised according to the following structure: + 
+        <Your chosen location> / Recording site / Year / Month / filename
+

For example: D:\Vögel\56075, Karthäuserhofgelände\2024\Juli\20240704_0130.flac

+
Library FormatThe library can store your audio in a lossy or lossless format. +
    +
  • + Lossy: Audio is converted to Ogg Vorbis. This is the recommended option. The audio will be virtually indistinguishable from the original in the application - both to the model and to the human ear. + The file size will be dramatically reduced - to the extent that a full 24 hours of recordings will require about 1 Gigabyte of storage. +
    • +
  • + Lossless: If the lossless option is selected, the audio fidelity (bitrate, channels and sample rate) will be identical to the original file - even WAV files. By using the FLAC format, the archived files will + have roughly half the size of their WAV equivalents. +
    • +
+
Spectrogram Preferences
ColourmapChoose a colour theme for the spectrogram display, or create your own. If you select 'custom', you will have the option to set the colours for peak, mid and quiet sounds according + to personal preference. You can also adjust the mid-point position: with a value of 0 or 1, the Spectrogram will be two-tone. Values in between will blend the three colours. + If you set the Mid colour the same as one of the others, you will be able to adjust the contrast in the spectrogram using Mid Position adjustments. +

In combination with audio filter adjustments, a custom colourmap allows you to enhance the contrast / visibility of calls using the colours of your choice.

+
Window FunctionA variety of windowing functions are available for the spectrogram display. Each has slightly different characteristics, so changing the window function may also enhance the + appearance of the calls in the display. +
TimelineThe spectrogram timeline has two modes: +
    +
  • + Timecode displays the time elapsed from the beginning of the file. +
    • +
  • + Time of Day relies on file timestamps to determine the actual time of detected calls. + If a file's timestamp does not reflect the recording time, the reading will be inaccurate. + To edit the start time of the current file, set the desired date and time using the filename context menu. +
    • +
+
Advanced
Send Filtered Audio for Analysis + By default, the filter settings in the Audio Preferences are just used to modify the audio in the application interface. If you enable this setting, + the modified audio will be used for analysis. This is usually a bad idea, as it impairs the accuracy of the models in most situations. However, if your + recordings have a considerable amount of low frequency noise, it may be that modifying the audio using these filters will improve the models' ability to + make accurate detections. +
Context Mode This mode is only available for the Nocmig model. When enabled, the model wird das surrounding context when making predictions. This helps mitigate + against false positive detections. +
SNR filterThis setting is also only available for the Nocmig model. The SNR (Signal to Noise Ratio) filter can only be enabled when using the CPU backend. + When enabled, Chirpity will disregard audio segments with no distinct sound event. + A stronger signal is required for a prediction to be attempted when higher signal-to-noise values are set. + The purpose is to deliver significant speed gains, however, the setting may cause Chirpity to miss quieter, more distant calls. +
Debug mode + If you have encountered an issue while using Chirpity, you can enable debug mode. When you change this setting, you will need to quit and relaunch the application + for the change to take effect. When launched in debug mode, two windows will open, and each will print debugging information to a console on the right of the window. + This is extremely useful if you want to report a bug - as you can share a screenshot of any errors that appear (these will show up in red in the console). +
+ + + \ No newline at end of file diff --git a/Help/settings.fr.html b/Help/settings.fr.html new file mode 100644 index 0000000..aa1045f --- /dev/null +++ b/Help/settings.fr.html @@ -0,0 +1,315 @@ + + + + + Paramètres Chirpity Nocmig + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Détections
ModèleChoisissez le modèle à utiliser pour la détection. +
    +
  • + Nocmig utilise un modèle conçu pour analyser les appels nocturnes des oiseaux européens. +
    • +
  • + BirdNET utilise le modèle BirdNET développé par Stefan Kahl et al. du K. Lisa Yang Center for Conservation Bioacoustics. +
    • +
+
Seuil de ConfianceCe paramètre détermine le niveau de confiance minimal requis pour que le modèle rapporte une détection. + Augmenter la valeur réduit les faux positifs, mais augmente le risque de manquer des appels authentiques. + Note : Le seuil de confiance peut être ajusté à tout moment. Les résultats sont mis à jour même après l'analyse. +
Liste +

Oiseaux Locaux. Exclut les oiseaux peu susceptibles d’être présents dans votre région.

+

+ Avec cette option sélectionnée, un panneau supplémentaire apparaît. La valeur du Seuil représente la fréquence des listes eBird. + Par exemple, une valeur de 0,03 signifie que les oiseaux qui apparaissent dans moins de 3 % des listes eBird de la région sont exclus. + Si vous activez également "Utiliser la Semaine", cela exclut les oiseaux non prévus à l’époque où l’enregistrement a été réalisé - lu à partir du timestamp du fichier. +
Note : Actuellement, une limitation des filtres spécifiques à la semaine est que le résumé des détections utilisera la liste du dernier fichier d’un lot. + Par exemple, si vous analysez des fichiers des 22 avril, 23 octobre et 24 février dans un lot, la liste du 24 février sera utilisée pour le résumé. +

+

Oiseaux Nocturnes. Cette option exclut les détections de chants d’oiseaux et des appels d’espèces qui ne vocalisent pas habituellement la nuit. + Lorsque vous activez "Utiliser les paramètres de localisation" pour le modèle BirdNET, la liste comprendra les oiseaux de la liste Oiseaux Locaux, en excluant ceux + qui ne vocalisent pas la nuit. +

+

Oiseaux. Inclut les détections de tous les oiseaux, à l’exclusion des classes non aviaires.

+

Tout. En plus de toutes les espèces d’oiseaux, affiche les détections non aviaires, y compris les segments considérés comme du bruit.

+

Personnalisé. Fournissez une liste à utiliser par le modèle. Les listes personnalisées vous permettent de :

+
    +
  1. Définir précisément les détections que vous souhaitez que Chirpity rapporte
    2. +
  2. Rechercher uniquement une espèce spécifique
    3. +
+ Vous pouvez utiliser une liste personnalisée différente pour les modèles BirdNET et Nocmig. Le format pour une + liste personnalisée est "nom scientifique_nom commun", chaque espèce sur une nouvelle ligne. Il est important que la liste personnalisée utilise des étiquettes dans votre langue sélectionnée. + Vous pouvez exporter une liste de départ au format correct à partir du menu d’aide, en utilisant le lien "Quelles espèces sont détectées ?". +
Analyser uniquement les périodes nocturnes + Recherchez uniquement des détections la nuit. Chirpity calcule le début et la fin du crépuscule civil à partir des horodatages de vos fichiers audio, en fonction de la localisation que vous avez définie. + Si votre enregistrement couvre à la fois des périodes diurnes et nocturnes, les périodes diurnes seront ignorées. Si vous êtes uniquement intéressé par les appels nocturnes, cela présente deux avantages : +
    +
  1. Il y aura moins de faux positifs, car les deux modèles font plus d’erreurs lorsque de nombreuses espèces vocalisent en même temps (par exemple, pendant le chœur de l’aube).
    2. +
  2. L’analyse sera plus rapide, car elle ignore les périodes diurnes.
    3. +
+
Afficher le statut des espèces sur la Liste Rouge de l’IUCN dans le résumé + Affichez le statut global des espèces sur la Liste Rouge dans le tableau récapitulatif, tel qu’évalué par l’IUCN. Il existe 7 catégories : +
    +
  1. Préoccupation mineure (LC)
    2. +
  2. Quasi menacé (NT)
    3. +
  3. Vulnérable (VU)
    4. +
  4. En danger (EN)
    5. +
  5. En danger critique (CR)
    6. +
  6. Éteint à l’état sauvage
    7. +
  7. Éteint
    8. +
+ Si vous cliquez sur l’icône, cela ouvrira la page d’évaluation avec l’aire de répartition de l’espèce et les détails de l’évaluation dans votre navigateur web. Les données proviennent de : + + IUCN 2024. Liste Rouge de l’IUCN des Espèces Menacées. Version 2024-2. + +
Effacer le cache des appels de référence + Les métadonnées de comparaison des espèces, qui déterminent ce que vous voyez lorsque vous sélectionnez "Comparer avec les appels de référence" dans un menu contextuel, sont enregistrées sur le disque. + Si vous effacez le cache, les données enregistrées seront supprimées et les données des espèces seront redemandées sur le site Web de Xeno-Canto lors de votre prochaine demande de comparaison. + Cela peut être utile si un fichier de référence est supprimé du site ou si vous souhaitez mettre à jour la sélection des appels. +
Emplacement par défaut +
+ Le menu contextuel accessible à partir du nom de fichier contient des options pour modifier l'heure de début du fichier et l'emplacement d'enregistrement +
Menu contextuel pour le nom de fichier.
+
+ Définissez l’emplacement par défaut à utiliser pour calculer le début et la fin du crépuscule civil et également pour générer votre liste Oiseaux Locaux. + Ce paramètre peut être remplacé pour des fichiers individuels en attribuant un emplacement personnalisé dans le menu contextuel du nom de fichier. +
Système
Langue + Choisissez la langue pour les étiquettes des "noms communs", lorsque des traductions existent. Les étiquettes traduites proviennent de eBird. +
BackendCette option n'est visible que lors de l'utilisation du modèle Nocmig. Choisissez comment les prédictions sont calculées sur votre ordinateur. +
    +
  • + CPU utilisera le processeur pour les prédictions, ce qui est généralement plus rapide si vous avez une carte graphique intégrée. +
    • +
  • + WebGL utilisera la carte graphique. Si vous avez une carte graphique dédiée, cela sera généralement + beaucoup plus rapide. +
    • +
  • + WebGPU utilise une nouvelle technologie et utilise également les processeurs graphiques. Cela ressemble à WebGL, + mais est connu pour accélérer les temps de traitement sur les Macs, contrairement à WebGL. WebGPU consomme plus de mémoire que WebGL, et + si les valeurs de threads ou de taille de lot sont trop élevées, Chirpity peut manquer de mémoire. Dans ce cas, + le spectrogramme peut disparaître de l'écran ou l'application peut devenir non réactive. Utilisez une valeur inférieure pour les threads et/ou la taille de lot. +
    • +
+
ThreadsCe paramètre vous permet de sélectionner le nombre de threads à utiliser pour les prédictions. + Lors de l'utilisation du backend CPU, le nombre de threads est automatiquement réglé pour correspondre au nombre de cœurs CPU disponibles sur l'ordinateur. + Par défaut, le backend GPU utilise deux threads. Ajuster ce paramètre, ainsi que la taille de lot, peut augmenter la vitesse de traitement. +
Taille du lotCe paramètre détermine le nombre de segments audio de trois secondes à traiter pour chaque demande au modèle. + Des valeurs plus grandes peuvent accélérer le traitement, en particulier pour les fichiers longs, bien que des valeurs trop grandes réduisent la vitesse de traitement et puissent même entraîner des problèmes de mémoire. + La valeur optimale peut également varier en fonction du format de fichier analysé. +
Activer les notifications de fin d'analyse + Cela est utile si vous souhaitez exécuter l’analyse Chirpity en arrière-plan + ou avec la fenêtre minimisée. Une alerte système vous avertira lorsque l'analyse sera terminée. +
Préférences audio
Réglage du volumeSi vos enregistrements sont très silencieux, vous pouvez augmenter le volume en augmentant le gain. Le volume audio peut être augmenté jusqu'à 50 décibels.
Normaliser l'audio + Si activé, les niveaux dans le fichier audio seront normalisés. En termes simples, cela maximise la plage de volume entre les sons les plus silencieux et les plus forts. + L'effet garantit que tout l'audio joue à un volume similaire, et les vocalisations dans les fichiers audio silencieux seront plus faciles à entendre. +
Filtre Passe-Haut Ce filtre supprime les sons en dessous de la fréquence sélectionnée. + Conçu principalement pour les enregistrements effectués en milieu urbain, il réduit les bourdonnements de basse fréquence. +
Filtre Étagère Basse Ce filtre est similaire au filtre Passe-Haut. Cependant, au lieu de simplement supprimer le bruit, il réduit le volume des bruits de basse fréquence. + L'avantage d'un filtre Étagère Basse est que vous pouvez réduire les bourdonnements tout en pouvant entendre les espèces ayant des appels de basse fréquence. + Vous pouvez régler la fréquence seuil et la force du filtre à appliquer (Atténuation). +
Exportation Audio
Format et Débit BinaireChoisissez entre des formats avec ou sans perte pour les extraits audio que vous exportez depuis Chirpity.
Convertir en monoCombinez les pistes stéréo en un seul canal. Cela réduira la taille des fichiers.
DécorateursCes deux options peuvent être utilisées pour : +
    +
  • + Remplir : Ajoutez deux secondes au début et à la fin de l'audio sélectionné pour l'exportation. +
    • +
  • + Fondu : Appliquez un effet de fondu au début et à la fin de vos fichiers exportés. +
    • +
+
Bibliothèque Audio
Ajouter à la bibliothèque lors de la sauvegardeLorsque vous enregistrez un ensemble de résultats dans la base de données d'archive, les fichiers audio associés seront convertis au format bibliothèque et sauvegardés dans l'emplacement d'archive.
Couper l’audio de la bibliothèque aux périodes nocturnesCette option coupe les sections de début ou de fin d’un fichier pour supprimer l’audio enregistré pendant la journée. Plus précisément, avant la fin du crépuscule civil le soir et après la fin du crépuscule civil le matin. + Ces périodes sont définies selon les paramètres de localisation du fichier.
Emplacement de la BibliothèqueSélectionnez un dossier dans lequel stocker la bibliothèque audio. Les fichiers seront organisés selon la structure suivante : + 
+        <Emplacement choisi> / Site d'enregistrement / Année / Mois / Nom du fichier
+

Par exemple : D:\Oiseaux\56075, Karthäuserhofgelände\2024\Juillet\20240704_0130.flac

+
Format de la BibliothèqueLa bibliothèque peut stocker votre audio dans un format avec ou sans perte. +
    +
  • + Avec perte : L’audio est converti en Ogg Vorbis. C'est l'option recommandée. L’audio sera pratiquement indistinguable de l’original dans l’application - tant pour le modèle que pour l’oreille humaine. + La taille du fichier sera considérablement réduite - à tel point que 24 heures complètes d'enregistrements nécessiteront environ 1 Go de stockage. +
    • +
  • + Sans perte : Si l'option sans perte est sélectionnée, la fidélité audio (débit, canaux et taux d'échantillonnage) sera identique au fichier d'origine - y compris les fichiers WAV. En utilisant le format FLAC, les fichiers archivés auront + une taille environ deux fois plus petite que leurs équivalents WAV. +
    • +
+
Préférences du Spectrogramme
Palette de CouleursChoisissez un thème de couleur pour l'affichage du spectrogramme ou créez votre propre palette. Si vous sélectionnez 'personnalisé', vous aurez la possibilité de définir les couleurs pour les sons forts, moyens et silencieux + selon vos préférences. Vous pouvez également ajuster la position du point central : avec une valeur de 0 ou 1, le spectrogramme sera bicolore. Les valeurs intermédiaires mélangeront les trois couleurs. + Si vous définissez la couleur centrale identique à l’une des autres, vous pourrez ajuster le contraste du spectrogramme à l’aide des réglages de la position centrale. +

En combinaison avec des ajustements de filtres audio, une palette de couleurs personnalisée vous permet d'améliorer le contraste/la visibilité des appels en utilisant les couleurs de votre choix.

+
Fonction de FenêtreUne variété de fonctions de fenêtre est disponible pour l'affichage du spectrogramme. Chacune a des caractéristiques légèrement différentes, donc changer la fonction de fenêtre peut également améliorer + l'apparence des appels dans l'affichage. +
ChronologieLa chronologie du spectrogramme dispose de deux modes : +
    +
  • + Code temporel affiche le temps écoulé depuis le début du fichier. +
    • +
  • + Heure du jour s'appuie sur les horodatages du fichier pour déterminer l'heure réelle des appels détectés. + Si l'horodatage d'un fichier ne reflète pas l'heure de l'enregistrement, la lecture sera incorrecte. + Pour modifier l'heure de début du fichier actuel, définissez la date et l'heure souhaitées à l'aide du menu contextuel du nom de fichier. +
    • +
+
Avancé
Envoyer l'audio filtré pour analyse + Par défaut, les paramètres de filtre dans les Préférences audio sont uniquement utilisés pour modifier l'audio dans l'interface de l'application. Si vous activez cette option, + l'audio modifié sera utilisé pour l'analyse. Cela est généralement déconseillé, car cela réduit la précision des modèles dans la plupart des situations. Cependant, si vos + enregistrements contiennent une quantité importante de bruit de basse fréquence, il est possible que la modification de l'audio à l'aide de ces filtres améliore la capacité des modèles à + effectuer des détections précises. +
Mode Contexte Ce mode est uniquement disponible pour le modèle Nocmig. Lorsqu'il est activé, le modèle utilise le contexte environnant lors des prédictions. Cela aide à réduire + les faux positifs. +
Filtre SNRCe paramètre est également uniquement disponible pour le modèle Nocmig. Le filtre SNR (Rapport Signal/Bruit) peut uniquement être activé lors de l'utilisation du backend CPU. + Lorsqu'il est activé, Chirpity ignore les segments audio sans événement sonore distinct. + Un signal plus fort est nécessaire pour tenter une prédiction lorsque des valeurs SNR plus élevées sont définies. + Ce paramètre vise à accélérer significativement le traitement. Cependant, il peut entraîner la perte d'appels plus faibles ou plus distants. +
Mode Débogage + Si vous avez rencontré un problème en utilisant Chirpity, vous pouvez activer le mode débogage. Lorsque vous changez ce paramètre, vous devez quitter et relancer l'application + pour que la modification prenne effet. Lorsqu'elle est lancée en mode débogage, deux fenêtres s'ouvrent, chacune affichant des informations de débogage dans une console sur le côté droit de la fenêtre. + Cela est extrêmement utile si vous souhaitez signaler un bug, car vous pouvez partager une capture d'écran de tous les erreurs qui apparaissent (celles-ci s'afficheront en rouge dans la console). +
+ + diff --git a/Help/settings.nl.html b/Help/settings.nl.html new file mode 100644 index 0000000..52e8cac --- /dev/null +++ b/Help/settings.nl.html @@ -0,0 +1,318 @@ + + + + + Chirpity Nocmig Settings + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Détections
ModèleChoisissez le modèle à utiliser pour la détection. +
    +
  • + Nocmig will use a model tuned for the nocturnal flight calls of European birds for analysis. +
    • +
  • + BirdNET will use the BirdNET model developed by Stefan Kahl et al. from the K. Lisa Yang Center for Conservation Bioacoustics +
    • +
+
Seuil de ConfianceCe paramètre détermine le niveau de confiance minimal requis pour que le modèle signale une détection. + Increasing the value reduces false positives but increases the risk of missing genuine calls. + N.B. Confidence can be adjusted at any time. The results are updated even after analysis. +
Lijste

Oiseaux Locaux. Exclude birds unlikely to be found in your location.

+

With this option selected, an additional panel appears. The Threshold value represents the eBird checklist frequency. + So, a value of 0.03 means birds that are predicted to occur in less than 3% of eBird checklists in the area are excluded. + If you also check "Use Week", it excludes birds not expected at the time of year the recording was made - which is read from the file's timestamp. +
N.B. Currently, a limitation to the week specific filters is that the summary of detections will use the + list from the latest file in a batch of files. + For example, if you analyse files from April 22, October 23 and February 24 in a batch, the list for February 24 will be used for the summary. +

+

Nocturnal Oiseaux. This will exclude detections of bird song and the calls of species which do not typically vocalise during the night. + When you enable "Use Location settings" for the BirdNET model, the list will comprise the birds on the Oiseaux Locaux list, excluding those + which do not call at night.

+

Oiseaux. Include detections for all birds, excluding non-avian classes. +

+

Tout. In addition to all bird species, show non-bird detections including segments that were considered noise.

+

Personnalisé. Provide a list for the model to use. Personnalisé lists allow you to:

+
    +
  1. Precisely define the detections you would like Chirpity to report on
    2. +
  2. Search only for a specific species
    3. +
+ You may use a different custom list for both BirdNET and Nocmig models. The format for a + custom list file is "scientific name_common name", with each species on a new line. Importantly, the custom list needs to use labels in your selected language. You can export a starting + list in the correct format from the Help menu, using the "What species are detected?" link. +
Analyser uniquement les périodes nocturnes + Only search for detections at night. Chirpity calculates the end and start of civil twilight from the timestamps of your audio files using the location you have set. If your recording + spans both daylight and night time periods, the day periods will be skipped. If you are only interested in nocturnal calls this has two benefits: +
    +
  1. There will be fewer false positives, since both models make more mistakes when many species vocalise at once (e.g. during the dawn chorus)
    2. +
  2. The analysis will complete faster, since it ignores any daytime period
    3. +
+
Show Species IUCN Red Lijste status in the summary + Display the global Red Lijste status of the species in the summary table, as assessed by the IUCN There are 7 categories: +
    +
  1. Least Concern (LC)
    2. +
  2. Near Threatened (NT)
    3. +
  3. Vulnerable (VU)
    4. +
  4. Endangered (EN)
    5. +
  5. Critically ENdangered (CR)
    6. +
  6. Extinct in the Wild
    7. +
  7. Extinct
    8. +
+ If you click the icon, it will open the assessment page with the species' range and details of the assessment in your web browser. The data is sourced from: + + IUCN 2024. IUCN Red Lijste of Threatened Species. Version 2024-2 + +
Effacer le cache des appels de référence + The species' comparison metadata that determine what you see when you select "Compare with Reference Calls" on a context menu are saved to disk. If you clear the cache, + the saved data are deleted, and species data will be re-requested from the Xeno-Canto website when you next ask for a comparison. This may be useful if a reference file + is removed from the site, or if you would like to update the selection of calls. +
Emplacement par Défaut +
+ The context menu accessed from the filename contains options to amend the file's start time and the recording location +
File name context menu.
+
+ Set the default location to use for calculating the start and end of civil twilight and also generating your Oiseaux Locaux list. This setting can be overridden for individual + files by assigning a custom location in the filename context menu. +
System
Langue + Choose the language for the "common name" labels, where translations exist. The translated labels have been obtained from eBird. +
BackendThis option is only visible when using the Nocmig model. Choose how predictions are calculated on you computer. +
    +
  • + CPU will use the CPU for predictions, this is generally faster if you have an integrated + graphics card. +
    • +
  • + WebGL will use the graphics card. If you have a dedicated graphics card, this will typically be + significantly faster. +
    • +
  • + WebGPU This employs a new technology and will also use graphics processors. It is similar to WebGL, + but is known to speed up the processing times on Macs, unlike WebGL. WebGPU uses more memory than WebGL, and + if the threads or batch size values are set too high, Chirpity may run out of memory. If this happens + the spectrogram may disappear from the screen or the app may become unresponsive. Use a lower value for threads and / or batch size. +
    • +
+
ThreadsThis setting allows you to select the number of threads to use for predictions. + When using the CPU backend, the number of threads is automatically set to match the number of CPU cores available on the computer. + By default, the GPU backend uses two threads. Adjusting this setting, along with the batch size, can increase the speed of processing. +
Taille du LotThis setting determines the number of three-second audio chunks to process for each model request. + Larger values may lead to faster processing, especially on long files, although too large a value will reduce the processing speed and may even cause memory issues. + The optimal value may also differ depending on the file format being analysed. +
Activer les notifications de fin d'analyse + This is useful if you would like to run Chirpity analysis in the background, + or with the window minimised. A system alert will notify you when the analysis completes. +
Audio Preferences
Ajustement du VolumeIf your recordings are very quiet, you can increase the loudness of the audio by increasing the gain. The volume of audio can be increased by up to 50 decibels.
Normaliser l'audio + If enabled, the levels in the audio file will be normalised. In simple terms, this will maximise the volume range between the quietest and loudest sounds. + The effect will ensure all audio plays at a similar volume, and vocalisations in quiet audio files will be easier to hear. +
Filtre Passe-Haut This filter removes the sound below the frequency selected. + Intended primarily for recordings made in in urban settings, this suppresses low-frequency hum. +
Filtre Passe-Bas This filter is similar to the Filtre Passe-Haut. However, rather than simply remove noise, it reduces the volume of low frequency noise. + The advantage of a low shelf filter is that you can reduce the hum, whilst still being able to hear species which have low frequency calls. + You can set the threshold frequency, and the strength of the filter to apply (Attenuation). +
Audio Export
Format et DébitChoose from lossless or lossy formats for the audio clips you export from Chirpity.
Convertir en monoCombine stereo tracks to a single channel. This will result in smaller file sizes.
DécorateursThese two options can be used to: +
    +
  • + Pad: Add two seconds to the beginning and end of the audio selected for export +
    • +
  • + Fade: Apply a fade effect to the beginning and end of your exported files. +
    • +
+
Audio Library
Ajouter à la Bibliothèque lors de l'EnregistrementWhen you save a set of results to the archive database, the associated audio files will be converted to the library format and saved in the archive location.
Couper l'audio de la bibliothèque aux périodes nocturnesThis will crop the start or end sections of a file to remove the audio that was recorded during daylight. Specifically, before the end of Civil Twilight in the evening and after the end of Civil Twilight in the morning. + Those periods are set according to the file's location settings.
Emplacement de la BibliothèqueSelect a folder in which to store the audio library. The files will be organised according to the following structure: + 
+        <Your chosen location> / Recording site / Year / Month / filename
+

For example: D:\Vögel\56075, Karthäuserhofgelände\2024\Juli\20240704_0130.flac

+
Format de la BibliothèqueThe library can store your audio in a lossy or lossless format. +
    +
  • + Lossy: Audio is converted to Ogg Vorbis. This is the recommended option. The audio will be virtually indistinguishable from the original in the application - both to the model and to the human ear. + The file size will be dramatically reduced - to the extent that a full 24 hours of recordings will require about 1 Gigabyte of storage. +
    • +
  • + Lossless: If the lossless option is selected, the audio fidelity (bitrate, channels and sample rate) will be identical to the original file - even WAV files. By using the FLAC format, the archived files will + have roughly half the size of their WAV equivalents. +
    • +
+
Préférences du Spectrogramme
Carte de CouleursChoose a colour theme for the spectrogram display, or create your own. If you select 'custom', you will have the option to set the colours for peak, mid and quiet sounds according + to personal preference. You can also adjust the mid-point position: with a value of 0 or 1, the Spectrogram will be two-tone. Values in between will blend the three colours. + If you set the Mid colour the same as one of the others, you will be able to adjust the contrast in the spectrogram using Mid Position adjustments. +

In combination with audio filter adjustments, a custom colourmap allows you to enhance the contrast / visibility of calls using the colours of your choice.

+
Fonction de FenêtreA variety of windowing functions are available for the spectrogram display. Each has slightly different characteristics, so changing the window function may also enhance the + appearance of the calls in the display. +
ChronologieThe spectrogram timeline has two modes: +
    +
  • + Timecode displays the time elapsed from the beginning of the file. +
    • +
  • + Time of Day relies on file timestamps to determine the actual time of detected calls. + If a file's timestamp does not reflect the recording time, the reading will be inaccurate. + To edit the start time of the current file, set the desired date and time using the filename context menu. +
    • +
+
Avancé
Envoyer l'Audio Filtré pour Analyse + By default, the filter settings in the Audio Preferences are just used to modify the audio in the application interface. If you enable this setting, + the modified audio will be used for analysis. This is usually a bad idea, as it impairs the accuracy of the models in most situations. However, if your + recordings have a considerable amount of low frequency noise, it may be that modifying the audio using these filters will improve the models' ability to + make accurate detections. +
Mode Contexte This mode is only available for the Nocmig model. When enabled, the model will use the surrounding context when making predictions. This helps mitigate + against false positive detections. +
Filtre SNRThis setting is also only available for the Nocmig model. The SNR (Signal to Noise Ratio) filter can only be enabled when using the CPU backend. + When enabled, Chirpity will disregard audio segments with no distinct sound event. + A stronger signal is required for a prediction to be attempted when higher signal-to-noise values are set. + The purpose is to deliver significant speed gains, however, the setting may cause Chirpity to miss quieter, more distant calls. +
Mode Debug + If you have encountered an issue while using Chirpity, you can enable debug mode. When you change this setting, you will need to quit and relaunch the application + for the change to take effect. When launched in debug mode, two windows will open, and each will print debugging information to a console on the right of the window. + This is extremely useful if you want to report a bug - as you can share a screenshot of any errors that appear (these will show up in red in the console). +
+ + + \ No newline at end of file diff --git a/Help/settings.pt.html b/Help/settings.pt.html new file mode 100644 index 0000000..52e8cac --- /dev/null +++ b/Help/settings.pt.html @@ -0,0 +1,318 @@ + + + + + Chirpity Nocmig Settings + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Détections
ModèleChoisissez le modèle à utiliser pour la détection. +
    +
  • + Nocmig will use a model tuned for the nocturnal flight calls of European birds for analysis. +
    • +
  • + BirdNET will use the BirdNET model developed by Stefan Kahl et al. from the K. Lisa Yang Center for Conservation Bioacoustics +
    • +
+
Seuil de ConfianceCe paramètre détermine le niveau de confiance minimal requis pour que le modèle signale une détection. + Increasing the value reduces false positives but increases the risk of missing genuine calls. + N.B. Confidence can be adjusted at any time. The results are updated even after analysis. +
Lijste

Oiseaux Locaux. Exclude birds unlikely to be found in your location.

+

With this option selected, an additional panel appears. The Threshold value represents the eBird checklist frequency. + So, a value of 0.03 means birds that are predicted to occur in less than 3% of eBird checklists in the area are excluded. + If you also check "Use Week", it excludes birds not expected at the time of year the recording was made - which is read from the file's timestamp. +
N.B. Currently, a limitation to the week specific filters is that the summary of detections will use the + list from the latest file in a batch of files. + For example, if you analyse files from April 22, October 23 and February 24 in a batch, the list for February 24 will be used for the summary. +

+

Nocturnal Oiseaux. This will exclude detections of bird song and the calls of species which do not typically vocalise during the night. + When you enable "Use Location settings" for the BirdNET model, the list will comprise the birds on the Oiseaux Locaux list, excluding those + which do not call at night.

+

Oiseaux. Include detections for all birds, excluding non-avian classes. +

+

Tout. In addition to all bird species, show non-bird detections including segments that were considered noise.

+

Personnalisé. Provide a list for the model to use. Personnalisé lists allow you to:

+
    +
  1. Precisely define the detections you would like Chirpity to report on
    2. +
  2. Search only for a specific species
    3. +
+ You may use a different custom list for both BirdNET and Nocmig models. The format for a + custom list file is "scientific name_common name", with each species on a new line. Importantly, the custom list needs to use labels in your selected language. You can export a starting + list in the correct format from the Help menu, using the "What species are detected?" link. +
Analyser uniquement les périodes nocturnes + Only search for detections at night. Chirpity calculates the end and start of civil twilight from the timestamps of your audio files using the location you have set. If your recording + spans both daylight and night time periods, the day periods will be skipped. If you are only interested in nocturnal calls this has two benefits: +
    +
  1. There will be fewer false positives, since both models make more mistakes when many species vocalise at once (e.g. during the dawn chorus)
    2. +
  2. The analysis will complete faster, since it ignores any daytime period
    3. +
+
Show Species IUCN Red Lijste status in the summary + Display the global Red Lijste status of the species in the summary table, as assessed by the IUCN There are 7 categories: +
    +
  1. Least Concern (LC)
    2. +
  2. Near Threatened (NT)
    3. +
  3. Vulnerable (VU)
    4. +
  4. Endangered (EN)
    5. +
  5. Critically ENdangered (CR)
    6. +
  6. Extinct in the Wild
    7. +
  7. Extinct
    8. +
+ If you click the icon, it will open the assessment page with the species' range and details of the assessment in your web browser. The data is sourced from: + + IUCN 2024. IUCN Red Lijste of Threatened Species. Version 2024-2 + +
Effacer le cache des appels de référence + The species' comparison metadata that determine what you see when you select "Compare with Reference Calls" on a context menu are saved to disk. If you clear the cache, + the saved data are deleted, and species data will be re-requested from the Xeno-Canto website when you next ask for a comparison. This may be useful if a reference file + is removed from the site, or if you would like to update the selection of calls. +
Emplacement par Défaut +
+ The context menu accessed from the filename contains options to amend the file's start time and the recording location +
File name context menu.
+
+ Set the default location to use for calculating the start and end of civil twilight and also generating your Oiseaux Locaux list. This setting can be overridden for individual + files by assigning a custom location in the filename context menu. +
System
Langue + Choose the language for the "common name" labels, where translations exist. The translated labels have been obtained from eBird. +
BackendThis option is only visible when using the Nocmig model. Choose how predictions are calculated on you computer. +
    +
  • + CPU will use the CPU for predictions, this is generally faster if you have an integrated + graphics card. +
    • +
  • + WebGL will use the graphics card. If you have a dedicated graphics card, this will typically be + significantly faster. +
    • +
  • + WebGPU This employs a new technology and will also use graphics processors. It is similar to WebGL, + but is known to speed up the processing times on Macs, unlike WebGL. WebGPU uses more memory than WebGL, and + if the threads or batch size values are set too high, Chirpity may run out of memory. If this happens + the spectrogram may disappear from the screen or the app may become unresponsive. Use a lower value for threads and / or batch size. +
    • +
+
ThreadsThis setting allows you to select the number of threads to use for predictions. + When using the CPU backend, the number of threads is automatically set to match the number of CPU cores available on the computer. + By default, the GPU backend uses two threads. Adjusting this setting, along with the batch size, can increase the speed of processing. +
Taille du LotThis setting determines the number of three-second audio chunks to process for each model request. + Larger values may lead to faster processing, especially on long files, although too large a value will reduce the processing speed and may even cause memory issues. + The optimal value may also differ depending on the file format being analysed. +
Activer les notifications de fin d'analyse + This is useful if you would like to run Chirpity analysis in the background, + or with the window minimised. A system alert will notify you when the analysis completes. +
Audio Preferences
Ajustement du VolumeIf your recordings are very quiet, you can increase the loudness of the audio by increasing the gain. The volume of audio can be increased by up to 50 decibels.
Normaliser l'audio + If enabled, the levels in the audio file will be normalised. In simple terms, this will maximise the volume range between the quietest and loudest sounds. + The effect will ensure all audio plays at a similar volume, and vocalisations in quiet audio files will be easier to hear. +
Filtre Passe-Haut This filter removes the sound below the frequency selected. + Intended primarily for recordings made in in urban settings, this suppresses low-frequency hum. +
Filtre Passe-Bas This filter is similar to the Filtre Passe-Haut. However, rather than simply remove noise, it reduces the volume of low frequency noise. + The advantage of a low shelf filter is that you can reduce the hum, whilst still being able to hear species which have low frequency calls. + You can set the threshold frequency, and the strength of the filter to apply (Attenuation). +
Audio Export
Format et DébitChoose from lossless or lossy formats for the audio clips you export from Chirpity.
Convertir en monoCombine stereo tracks to a single channel. This will result in smaller file sizes.
DécorateursThese two options can be used to: +
    +
  • + Pad: Add two seconds to the beginning and end of the audio selected for export +
    • +
  • + Fade: Apply a fade effect to the beginning and end of your exported files. +
    • +
+
Audio Library
Ajouter à la Bibliothèque lors de l'EnregistrementWhen you save a set of results to the archive database, the associated audio files will be converted to the library format and saved in the archive location.
Couper l'audio de la bibliothèque aux périodes nocturnesThis will crop the start or end sections of a file to remove the audio that was recorded during daylight. Specifically, before the end of Civil Twilight in the evening and after the end of Civil Twilight in the morning. + Those periods are set according to the file's location settings.
Emplacement de la BibliothèqueSelect a folder in which to store the audio library. The files will be organised according to the following structure: + 
+        <Your chosen location> / Recording site / Year / Month / filename
+

For example: D:\Vögel\56075, Karthäuserhofgelände\2024\Juli\20240704_0130.flac

+
Format de la BibliothèqueThe library can store your audio in a lossy or lossless format. +
    +
  • + Lossy: Audio is converted to Ogg Vorbis. This is the recommended option. The audio will be virtually indistinguishable from the original in the application - both to the model and to the human ear. + The file size will be dramatically reduced - to the extent that a full 24 hours of recordings will require about 1 Gigabyte of storage. +
    • +
  • + Lossless: If the lossless option is selected, the audio fidelity (bitrate, channels and sample rate) will be identical to the original file - even WAV files. By using the FLAC format, the archived files will + have roughly half the size of their WAV equivalents. +
    • +
+
Préférences du Spectrogramme
Carte de CouleursChoose a colour theme for the spectrogram display, or create your own. If you select 'custom', you will have the option to set the colours for peak, mid and quiet sounds according + to personal preference. You can also adjust the mid-point position: with a value of 0 or 1, the Spectrogram will be two-tone. Values in between will blend the three colours. + If you set the Mid colour the same as one of the others, you will be able to adjust the contrast in the spectrogram using Mid Position adjustments. +

In combination with audio filter adjustments, a custom colourmap allows you to enhance the contrast / visibility of calls using the colours of your choice.

+
Fonction de FenêtreA variety of windowing functions are available for the spectrogram display. Each has slightly different characteristics, so changing the window function may also enhance the + appearance of the calls in the display. +
ChronologieThe spectrogram timeline has two modes: +
    +
  • + Timecode displays the time elapsed from the beginning of the file. +
    • +
  • + Time of Day relies on file timestamps to determine the actual time of detected calls. + If a file's timestamp does not reflect the recording time, the reading will be inaccurate. + To edit the start time of the current file, set the desired date and time using the filename context menu. +
    • +
+
Avancé
Envoyer l'Audio Filtré pour Analyse + By default, the filter settings in the Audio Preferences are just used to modify the audio in the application interface. If you enable this setting, + the modified audio will be used for analysis. This is usually a bad idea, as it impairs the accuracy of the models in most situations. However, if your + recordings have a considerable amount of low frequency noise, it may be that modifying the audio using these filters will improve the models' ability to + make accurate detections. +
Mode Contexte This mode is only available for the Nocmig model. When enabled, the model will use the surrounding context when making predictions. This helps mitigate + against false positive detections. +
Filtre SNRThis setting is also only available for the Nocmig model. The SNR (Signal to Noise Ratio) filter can only be enabled when using the CPU backend. + When enabled, Chirpity will disregard audio segments with no distinct sound event. + A stronger signal is required for a prediction to be attempted when higher signal-to-noise values are set. + The purpose is to deliver significant speed gains, however, the setting may cause Chirpity to miss quieter, more distant calls. +
Mode Debug + If you have encountered an issue while using Chirpity, you can enable debug mode. When you change this setting, you will need to quit and relaunch the application + for the change to take effect. When launched in debug mode, two windows will open, and each will print debugging information to a console on the right of the window. + This is extremely useful if you want to report a bug - as you can share a screenshot of any errors that appear (these will show up in red in the console). +
+ + + \ No newline at end of file diff --git a/Help/usage.es.html b/Help/usage.es.html index 36e6b9c..09277fc 100644 --- a/Help/usage.es.html +++ b/Help/usage.es.html @@ -7,10 +7,10 @@ -

Flujo de trabajo sugerido

+

Flujo de trabajo recomendado

    -
  1. Abre un archivo de audio, varios archivos o una carpeta de audios. Hay tres maneras de hacerlo: +
  2. Abra un archivo de audio, varios archivos o una carpeta de audios. Hay tres maneras de hacerlo:
    1. Arrastra y suelta los archivos o carpetas en la aplicación Chirpity Nocmig
    2. En la aplicación Chirpity Nocmig, selecciona Archivo > Abrir archivo de audio o presiona Ctrl-O y selecciona los archivos para abrir
      3. diff --git a/I18n/community.de.json b/I18n/community.de.json new file mode 100644 index 0000000..ab5e14f --- /dev/null +++ b/I18n/community.de.json @@ -0,0 +1,12 @@ +{ + "heading": [ + "Möchten Sie Hilfe?" + ], + "content": [ + "Es gibt eine WhatsApp-Community für Chirpity-Benutzer. Dort können Sie Probleme teilen, neue Funktionen vorschlagen und Artenbestimmungen in Gruppenchats diskutieren. Wenn Sie der Community beitreten möchten, drücken Sie einfach diesen Button, um zur WhatsApp-Einladungsseite zu gelangen:" + ], + "join-whatsapp": [ + "Der Chirpity WhatsApp-Community beitreten" + ] + } + \ No newline at end of file diff --git a/I18n/i18n.zip b/I18n/i18n.zip deleted file mode 100644 index 1544f42881d178ec1480cd31779dd8c4c682bff3..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 63946 zcma&sQ;=j^*e>d_ZQE9tZ5v&-ZQHhOSC?(uwr#skuYJym*n9u$V&#my%8a~t-Z8&9 z@{J)c1q^}$0P*+wQ|+J*@c+GP0096H0SEvLZLO`HZOonA>5TO0EF5iZRFq)>z&)C{ zO#lD4vl}!3Ajs3-@BjT7008Sa8QB4TnBW(lA*B^c@KZ3t_Fa%#ewwUdiA~I`xhQ01 z!Wbd?`N+MXTo->9MV7>qlbfy=)Y|KGtT~&mp$6TrJq4l+L>`%_Z#&*0r;cYn`lYIE zc#A3*g&TH5-jLFqs?Fj?b8j*SN`H$+pAEdbGuTsKn|3-_*@tZR7WhAf}FJx<_CFl(hPz@?v)uP?m-aY~RpyoZ;TYXV3F% z`UqQ{Q$&(y?QLd@9PhS#Kdht87}dP0y7E%s5O>TsZP0&h{_&LmSLv9)pZuTF#y0;+ z+E~6r7MKy?@&UE5QizuTR&X{53q>TXzioqyt8F#KHI+i}lgm=-5DH3a`-z+Tp$mvp z&IZE%y!o|u4(+4p!B10ha!>k}dJX^{W|*Yl+LBEYDVi zo&bdUOtX$h#U}h3)B^{db@#7^&eYqUBC`7TH_(49(Vvmvul_Evf5iUXa_E0rI{rsX zE?Hm!m@7vp%L`I)@jwJvQY3K{#9-IqHD{K)sAJ?tL9L(|V0n80l4k?AlswpihOh00 z+?@xgJWL2$sU0i5%4)c7-W$mG1g7CLT)+_nMjcit-ewiOPNRfQ2phW~B8Xx9h*}Zo zlwJMatQtW-Kg7N_5K`2rs^EeZ;E`T;tpRI7sd9_tG!DVAZbzryML{Ya5nPFUVzO0K z#!Hd4!t2x8`n3ww?YchXq6tb+rmA0KUaQK~vm;2e3K|2rLS()C=`iJvF-@SZ&jQf6 z766W16@CF(ma#%3|&uEgcRcrhgCkZhdU&!QbC?G4tL5Y82w0x zk5rE{LX*n?wMuDNB)Lh}w3;4wmu-l@n35560Px2PE2;Qee=C^|c%`k~0saXtQyLVhOJ0 zztP|EFb&~(GFwkYPT(JsIOec7tB`dY1B;esF+&-AXV&|bd=p%%OR{1QB1G?^!Oay8C>m2P8VkDzT4+{a3BJR> zxV_`J`5W6OUb{QB1WKl7_YPl7twdrOU6!ZE>E~|%;1GLw?!8QZ!}y2&-vep+A2)%G z)qe`aT7Kf6(cSq%Ey~mK<)C(IGpHqHcfMyPKKsubwG@V`BY7mdTmD?H--hMAa=E!xSQYIiF20nW{eu3TPrhF=8b#VB`*ob0 z^h`MN!C8s2_VpC=s^(I51uh#(!_$Gwg)X<3m7AaqUD|M>Gm=IA_HAgEvBz>gP`mbV ziT;m0EUPvF?qA)1g#X=s^MA}72j~A}@2n^z+b@r>?HPf-dFoH#nOhIbME9I5L8o-BDI^n`*sc+y0IBoy>$+&5t?q7<>|h1V>Q})6>Rb z&B2y)u!DW=Fe)T@2V(}*l?03eNO5Mkk~g2#bpCx6eu9gM;_0So zR-eGNZ&6^4AUH)d;sP25si`C#OhMD z_x&3q^rE%6I~9ppAi`^El)g8#*W#0u3Ah&+6I6!8H1)jx*g$a3K^4#3 zGBMYf(k!Z(TxFw&V#Rudr!!G!>*2X(;^W4NTTy!q=_l3JHI&uElH-%l2l$_S?M_`2 z{QqYAPm2GZZ~p(dw;WymQ@+-5JT?Oie|Mi*`-F}Oz($l2V5h73LZJn~hI6d6>1alj zBJz){XQ&>4x>Zhbxy62HP_g(sp1RO_YfKz-CT&9<(a(41QV=j|c5$B#oQw zORRyLrrw$EZo8fJ$DYQmjtEEmp0H=^iVhG=L5=teOBkXGu#rn-{V27%>&d~mhpG}) zU0Zt%t=LdRai;Z?vv(M!!vxJ-o#oI)Sb%3u5ryvMSZ@ScX2b(PHjR|GRe<+Jd}_Ra z5hI!gZw{`YYa>{7Ko&cLvD+Q6QtJFBZV-eptqTza0p;YFz}F_e7H~KWQBZdTp004{ z3G#QW-C5!+{ZYgsnCa-};X6zQB%bH$-vtw}9JIwh3$FCZ6Mv2cS^9XEy!{gKc>U+B zu!6)k%O@HtqTZq_>ClJwcIJ)^x)Pm+5AdG^Y{4!Si2oY@&^ z7}5R5t^bf(U(5z8Le~p5Sa&0+)=Q35d@E!-e`b-)Y|-g6zl32pg9sE!lTP7j&&!m2 zeUkuPa->V0-F`BAmOJy!-Q5d41fTCY`P?vTrs<*BN%Mm)6W_7-ZMyexA#6`=UA}HB z-0R8GLoVC6#<7qo)0!~Wbw{#*)7^N}>V_NCw)zYUdm=pHa}c~_cKC9OvP6Eq zBo4lGbDxr7{hFd}r+HicxpbT1dg$pI$NQWz3fv=^M`p*apy&1$1nj&gRI_s@TKUT; zO9}&c(A;fO$Y_ZKu4Bp~pC-M9FzYXe^4bb~!O60%cbvxegk6}O^$1@Gatk2xbyrq6 z)?CWLWw*?G&N(oE#kl{Xdi~*1=O0?r$Qs~#^mRd zz;Pn)h##3qrErBanw_!JxK>=ES9@u#Yl^)#cO+(@eAd)oGUk_U>Kn*_)yk6#lhVde zHi)o7$+*Lzh6yh?IS}TuigBvy;B3!$N}-fe5E(9ynixI84&WCn_0J5Vl&X5amaB2Y zcF_kit&W=rwkQdsW@c2u;#i;}BFar|kV?cymun6Cx(#4tF{AA5`dBhLrYB{b3OhLP z*j4j>Qw(b4zbygH{0K}(r^+_t{i2)`9yzMCqmlkx!jh?0UZ2C(r9TDp&&-JnV&cca z+OAkufUbjPb80;;Ib8Y^Ro}ei8MYmN&GtbTq+D|BQ;13yrItwWyEPVIdTQ=8OwjPRY1cM;BG=?WF`9S^n zTC&#HU*nIJ^jh|iKTi65D>to4om&NYhxu{-%&BumMl=N#>#}Y~tQLJ~nr?>Q(zqLi zDXNb*g(2&pJjZnlr%ytgUuxRXYT9fh3_0OW{EV06_`Req5{#xj=oAt#@7KSSZ<}rPd}o? z`{z~mI}BqdI)ohBiJPT=rfN}>jI#|Sj1n7pwiIzxD(BHRlU?wHqx6()m;bu-nu@R% zu5y@yu!V-bxm*`Nivx9ztl>seh~8e%kD-4Gp}EzbkhXBUR`}^h?X1yc=sS=b7lUhe zei@s}ODP@MmQyTJ1=J~IX|kc&;E^JUV9M_(TSBLyrat_lnQTNZ_9)aWyOCTNb^Z6= zI%c5)UV@TrIqf}%Z(-X${8&r#>pV5L9`W?&9mHcb#$lKZ8 zMva;>=)iG3c7A>*WF>X-20D#J$Ul>-zfX%A%0%;jMtQv&*c>nwKYTG7D3Fw+ra)^$ z1n3wV(x!bpsICTJts2+(QBDGZv^s|V44A>Uny*sY??vX6RFvgz0vr*7*hH1^N+|{Z zu04g0=cWpV?8j0da#VE81>yrBP_H4FBDBM0_~zd6emrNt^RA<2b+Z?hk`i=C?vsIu zvz~C5$M};=u#UYNYxK^x*-$U4_Y;mYNF@3NP=

      (_7 z_ByMFvDH$Fg{ec+uw=2d_Ei-n^+`2%JNw6<2Z6YaZFd!?YPFj88zn@zpM_)9Bhh4; zmc?j=nr4-~Y(B51WtQpIiYeS8vtD#hRNUtco6L&k?sJU=GS4UXUpa3^h3FUKHq*Z8lJAU3rOJJ}XRT87Zt%q!+W&K;Nj z{&6G5bJtd+Pr2CYMlCoOCqGABIvn#QE(I1b7}6KRK_m)QDjTwFK%DPC?<=aY?R+js zwSNQBWi7d?bGS=Y$PnSa-IcMpBV=FL3>-8kP=Q9J%gf$utfe?ay^3sw`>aDC5&(sw zTUzOEky~82YACIwW1AV4**c7jUEI4t^WDMY4W#7t)30>BS`e#0JS)6nZh9@=;IG3L zPt6=MdRWKno{jF1q3!gM|C75g+lFlWahGckLX(ikelnM>;nw98T-{4{f2vXIS3bH30UR`cLFAto>plivj znK3oHNA;2h8&@rNbZGN~ibC}ub2lWwYXq=g-3FJO;o)I;^|XrAO0m4BSe+YLMqkD8 zZE!rKuMHq40_tQEvvVwQK%5{kkOPhj(p8(z5&>WQ3E$-X(-QHw_Yb2c7!mn4lr|k~ zfcTZ8p$l!(?--B1<>3@mb1Pk@0K{tcNaj0&;S$e}C5^33aM4>DjSSg5=q3{V3hQzE zSExeTDUB8m#IfpMz2#a9@k5%T)?vER9o>I7_vfGn7{00(3thTE>?lJo1}c_NcIIj& z%QC#6!x5JBi%-0^g-Q3&-c+V}^Qdip1;SR(3|r_F0)Y3Db=6?Z`(WzqTQ=>zFqSYf z=Sbxp&1DV|`Mgzo8y7Vkzt$a*y@IE&mP#xtVg0ojxLj#B1z!5}1&btpXCvnz`SmUH z@Xz60T~FESbcsOZUQ)?(&S-01X@!XpL_g%z0dd)P@Vcr5$q*qt<7ur0#cd|(3HM^E z>XxMT0FX)H4le~vysn>S_qTZ$XYOs1POtX^I(M*uKJE?E+#MWT&crV9 z>6GclfRfHGH#*QBc!?wu2#}@wf|yfahXbAm=nj00ROti-od~2b<1eeCf6JBsu7q1%S9c{^-tU@5 zH@pJw#h;^2%Os;aExZlo1^=B&l&@C1)r>ql13F5-xSwodk?tn2iC^!Jy9pl4tb|hX zyh~ULA%2{(t%U7VZ6NQG8KG+y6w!W;$A7PY7~tnZrL5HwT-q$qAdF^aj!BF*q}qd6 z5GC0AUc42XQY=~OaG;!mokU&#Ov3UJ;$ldFs-nLe& z<58dkugc5HXkc-VO4qiwUX$hM3Lvlq#)#knb);pBm7A5{VF{?x^tD)T%8@&Lmq;Q7N#3l%Ll^V1@=_7TUY5D^d$1oBv=hD@I+%!^ z(<4w^>yZ>Cl!$?nyJog9>o&i-3UVo3@*@q9u=G%vHd-Qbx~28pqC$mLlRH0lQ6@Lk zAi4(>PQ1aLV%c3D@GYQKD&&;E0I&Bt`x(AIVRBm3j~AmZGb`;OuHzGdut*58jZil7 zm#*LeAi6`KHL;$`KH4Kjn0$rIjyNH+-C2tacq>s5U{t?Lt#BS&IRcV31AVZ#QlE-z zsZHL|{!3xL@xF1o*K7USw=9kLa{=&hDP1%aS=O-+JVz_7xMKERP@gSzHf`Q50wcWxKo_`7|B5z5NUmW62 z!H}!j3_AqEl>4vOEFqQeC)Lz^cYbm$BB^V?{24+$mK%)tOoU_#ec(4fY4G2~!wB#_ z1-QCbgbPSx@9@Upg9~WJ5WVyD__r1j6{lsiJmQE~$iexRdeIw%A-%%d%q_kPrgde| z0GsDngCWaUj1>I@6+9 zrjxMYxM(Us7cRrVnJnX{6KbeUG+)r(Nxv3%^Q}x^#rbrIhI;jGy}-fbQfG2hXMQ7L zOr!L6-pS#s%;fND@J3qu7pW3Cg(#AQVV20fZsCHTfpYkiHOr0arR|7MhWH4zBR>Nx z>l=~vMxI_g*oBodVjipfy64!_)CxOD!o}wHg%^CB{UzL#f1)3e+6LH2xIm(OF*ZFY z9fpRr6r_NSG`|Q!tWQ83)TU;4t|vguN^7S4I${@7mpgM{6=;Eut$5g&h%)?a_An^Z zVxrO~zhj|cTv%tQebcgmU{}BcLz`ZYdru@;5SI*GF@T7vjpGW4W1w1m$!Bk8a*Gdk!6uQ@Z8-zcSN{gxg$_(mZ{gRlkh z-0mL;ztjfBeo^W}b1at4;B)tFsMPpI-Tj%&q`0?Ph9eaD9k1OcMw-|{dFeiWD(7?K^K};~l*0dKJLrfUO`iEx?lY^xm_!(U z7U?puD71$@l(PS2`1|bg^D?+)itp=4IxIP&a@d(EE=|=&kL`hHS;h&j-NXJt!psM% z8Z#7nOg>ykGU6>mn&`tV{WENcUwYiF^BBu_{ovPqVM4f{>9Gu2`i^*#c?!zNL@R$p zumB`qHkE&-js$D!Qr1)<_Z&5)k(UZH!If3dIJhHf7Aa2nk6|~?n)s?a!%Kn`eKI}Uv74KdGw3qM-2>U@#jJ#ADme@x9D}o{K2{{nS)Y1x1Sd=;VvYVV= z133-GJq_wNyE6Gn5Xde0GvY$@0f%8@FF>RamJPZtqfNwn;Izb8>pE=LvIR^)a3}Os`6YIknywTgP<=j)HsE8A zD?1;3(k(0GS#rJOX0xrSSJp6f`w=@rHE-`+JVmKgEK|BU2vRU=ZbC4HI#=l3g=#K3UC(~EFUa#gkKC~`;LatAK5)FA6~x*#2Ujuz+e$#V6ctwUR`a1a`57pbi12{r)QBIg{!RPX3Nfn8u;V@zln-J zF0M@=H^6FsNU(VWoVu=@=g3dJWdwu~G&8rrnpVRs#OsrMZ`hZIl?+~=7z@3Vs7CCI z9eP{8UMH%^x8-JwL`2oV?~*CCGx$Z=E)+G!|BOB?*^<+}JMuR-}@}J?fC2axk zSZ6VIhZ`>`qnE;yHWc*$Y~s_{m2`}=+ot`ARQnO@k=6EXzfqRtYuz}C8KTefzRZ0i z_xZlI-TA`*ryG8i>Ew3$TV?!HTKs#Rp-%sQ)EUPAshLmKn6klXL+rj$VgLfVBC=TP zfOKGh0k5Z7i!rV)jJMVc#93Xq#4~v}O}SjIA{(WRMbxAB0n7EuM4wO_$~BpZT9Ng0 z7Tq$c`(5>ZUTN@if$jf;n~fTL%4J{dNIcbbs==7(N|kBJ{70o({^|TQ{`F(dbE#R7 z&eOXz`Alw5BVDj*--^OXZB)nnGoG6EyBx}RFg=~#?Qm9AHJy~~AT_Q!PeRUF=b}cp zthlMmDsuYKAb(>@zTDxX{9G>f+a4+D<3!I>!Nr=sYWuRX=}Rv1DX&4bpx0Mr>!^Mh ztKqHxwEoZJ^TD&)64lna&eE^OUa2Yrw^#KJRcrsyv4qA#y_l}2Y4r6fCtvy$q-VD_ z1`7+IEhrjL=&B@hqsn!kdfj?}F55wEu?0qcWs9z_+A{N9tKvtmd>G6k=47j;U|1DT zBG%IVv+_8hvmsrojXxG^FOp71cI1y5PnlSOcI*>M_OvI^s~zD6nsu8#R_0TBHTcmr zPSh89K3!jCj7ZRM)`Ujd3(w=KbXc8@STiiR`IxYwr4J&xQjS#HtkY=v()CwU1Xl9< zMg^cWjV7s-MW}yNI=)C^4iZKh4VjZLq~3A3?i^~j`;y9);fU2u-?BvxB1neo;~8V7 zhCFqjutM!V7Z}oqV#IAuaX9O?8I$sqpcL}fmF_07r$YnP3N9_AOTinqVJ7gXm8r@W z_>f}Nx9B^gF&*Efsa2EGfIuXL5&m*k^wB7E*%=b%RyH4cAOQD`_q@=DGbjv!hP0+%6Ny?w#-QH{+dA ztKahRi==FbyZmvlc9KQ%C(1@2jd*2n;p*Srto|cYhuvzua-ldwnthcj00YGCGyF^jU44c^FyeD=txA3Hy{WF za}nDp96hmE>`N4z(pPRZmm+s*b=cYm8X~b=tIHXP(yxOl;@n0N&l3Dg~5Q2PnCBPrg1~_V!cFGcjZMY5M zv>^(V_eQ*X?VMgaLC5RZADXlq)~aH#F0?U|!%cioW4#REGnJXWa% z9$3e{4N~(t3%Q65D@}_wg0SjS(12=z%NYkgw1y!-#fk1trzZC&LYO0Gdpt?O8&P#j z!>b~dXJ;QKXD$N}S2@N`_a3VzFmrC+Dh#3FW9ej1UYYjS1FS2q>CO#mId841>y zMLWa|UkXP74GAW;I+agRWgW!NGTAAvnBpo*m|o>sirE?6|ILt!!WR`L?IYwPK!A?} zG00D6!72GPQQ;U)o3?H5uVn8<35|Nlk=!#=r9m`>8dd(t+d(1KbP3wXV<|-fr*jSj z-JsjRe?dy-NpU3qvS0BbJsjcX7}StnI@wOkL28J_<#jNB^enW%zMWkM_D6sO&_x{N z)`h?>uZQ?gpuHnF=|ZAw#guYvafLVW!4O$~0*=Cx-fu##qm3kAGnh&`nL09J@J=ui#4k_TD!qDX#21thsC0e8Gkg(jNd^dpC9;(g9KdO^{|ff2h$x= zLx6+yAsqBFtKRRg-k#KKAOkFzVB_SMb`Nh+`WEadcm%!p_7FcQ$)t$oh(xXW7&AI6 z$4o7^6s2Dud@D{5$Jl5mPELNOtpIM+xt)?|RbVAKvEHacaHnvi?y};cgi|;cQtdzb zW!qYdHH+7>`^Qk9asf!b$Q(q1IPlY5JAlse>#tYt8c!qKv`a1tPaxyB3(Br0p2}{R z|IvQ)j8~>z$eN({u9J@O_FAH)EbZCw8FUzl1-TQiZti*hM$CVch?9X5Lk3uiI?CIP zxYQtS4X zS%-?qa`&43%88-VGHOwcUoTKk<5yeBq@LsWW7{?#e(DgKczbzavwcd15R!)ixGeTN zw1wWgk{Z@w8XdSb2ql~RpsUa^yV55;h8=T5jsx%OZEqlS+ljEYP$gY4 z8r#0Y8@I0kIGeOhJW3X-*O2^R^#RBLKYT~I;!v9V$#LH`9B$N{efQ5PKPZW;CM5;2 zD8Ee?4Ii&^x%|T#kAWJ!K0Xm|0Gz$CvGz1?J?1vA#^rZ%cg54cow`u$=lsv(qt zE@mOv$iBQ}FkFnK668(MV84gWV-OfPKiKX<1=QO2^Z@(8nxpb;ApMgcR;j>GJjLp& zd;UG}5CB`851Wk@V3$|x)Qt$OqsmQUtahO%c@gZN-hOHD{ss1FJ4`_}&MivPcuqQC z(i*k})Nx3Q5FJ-LetRM1<49^lMN@Vd4pQwMqif~C>Urn!FWJvCLu<)G;fy^Xtw&oRRQ7`=K&Z#c`XOO;E+Xz$H;NtyZc@|E$j@Gd3-s}kkf)R8K1 z)*_O$`J6Z#B@=yCf*bx{u}MWpv*>3^*%k0I+Nl(}?I@VZU2O8< z-x0BBW5;+}?o{9Zmf5mKZKqxOY6A5t>*vW73l+$91Ka*qbOVHz=}t!&-jwY7vKu$seUyE;*jstYJ>4R6w#m2^Lk zx>Z+Z zcAUw_qX0vd3qR>N!z6+ILRNd%Kb=_Yl+!Xew20LBj*6VFC&~fdGqKrlS3sq%`hw?} z&MotNClvqCky_mYCI#~PWV-Nl0?nWHbV`zrg*c96o^joD9mQu(p2iqa=ED>fY`iGz zRK@59pWQqA`vqZsMQN(*@yE9y3m8?IOv$%|i%T>TQALQdOF8lp;KWG@%&GsA=)tuj zoS8f|d~N8F^=?+i0Hp(qfgLi+n6DO|!-NqEmM7hbj~5Mo%F$Dr=gi6@!RH>vj)z=$ z>p^0b_(gj+ycHMF)cLApem~iPagCgf30ygy)@kjA%Bc74w8EUt%TO|zOLOv#_r=WB zN8U^gWJkc&c&w9ys)hWJElP439}b&%`lEouM`)B^4Kp;eNn+EZEc)+Kw@E$cy@`81 zsMVCHa)OkyQpOEMh1hQf#?Fiti0%>dI&pcoW^2VL#Q)gKfyaBZh2?-QUO%bGFA9gC ze=DJj>Eiy#yfDfBnu>{JA3l9>m6@b`#7vu@82HNXi9$di{OEFGqfj>D%)!|6#Nn-i zpsq(nbqv$Xc}>aX=x@oBwUgTcu-iFRaHvWmW>!bxjULAS+9C{Yj(#p^Q~Xt~iAXq_ zQ4{CXCX)2jBdZAGr z1%!wEplLZgxMe3}9t>qhJGM$t9aArh^X4L%E`?YE;QXoyW($5!%3Us155(A@Hvz!J~&O%MNz?m{SEyq^8GLB8f%PLsMR zYW#9V)01?YBL66aXsQd(i&0MrhZi^WTwnIv=@A(aH+9#kj=-ip2YQyZadQ*hB1pL% zapR}|YES;*D5hv~{t9>u`wwkMh|RN0^jP5M6aRe;E_)>okCK$Df&>QY4$`J`Nma+M zXaQKZZjH*${q#@F(Z@Fn2>N+Uo|S_1>L%iie0G`TiL-^6DBgkt+B@Buty1hU3bJ-1Wq_T|KmOB%lc4yh2_moCX=vB?! zTlnJ84~%ZZeC65*sn{HInSvhMN`o(HaeC`M;;2`IbIYcy3b1ivwqV4G7Fn^u-R2Z< zLLOf#2fa^4(3shDp>MinkcvN=9`H_9iSFI`I~z|dPk4Bay(BJWrud2#JPUDpa%L}V zMXBlHe0dkNI5{NSFAG=s)B4%=#;~W+4|XnTatW#KG$wL?u4zn;xZoecpvE#Fg}x^h z(f8J8QwJYs*nYl`eg7$l2&~@Sh7kb({;7ffy&!U<{ND?r|9E&XS>wVMdj#=kyT^b! zPd?z@XoT`zfI(|C;;J>-S|q_5MSzdyCgo7|W&A1|gKYF?duQ_ULYz}Fu$SoOxalEB z{rjCf_(ztmLNYqLy?#Ofo@=S;T$L}apixLY)4=fj`%vr4?cUu`!&sxMN~THPqr$3G zwV74+tdBxlC1flLyU;?e;#!hN17QSWwTtd zr2>1oy=l5kn+^8+vpM)uG`h?E%q$z4nVxa*yeMK-rjRJUg;xt*kdi8 z8(nVtj`2=}{IjcS+-Q%2Z|g@tzMEZU>`&71YMaRia+vQ=i66mi9p~vQ*)AWjhXo_> z`&VPO=<`fA`msimz!2hh_?239zM8H{p5Lnsw_w)UP&4#%H z!KXaS5O5^=1}kHW*BwFHEB4bln$CLXjSMX5lsJRalBbgG^5`1OOjLnRN|&&1~12_pzp5lvJ{}xRYL0pcocAXj*D2LU>0EAwA;Fy=3V$^8;e2r z+nx)o1Nc3MZ(@VH=ie@G0E@%zc?+aM4mSFT>kbT9yzml|7-?NKY<8lz6cpG zHRbSgt(boqCHq|rNXT*)!PFJ0L2SEBJCH!`5VGLgZ_Wp@0*GkO^OO;_Q(ZvMIGkv+ zQcTDI2b3Kx?#9n7ZYb3IM9v2Alcr`s1<|B~s`IgT!?vNT?)1}MnmK3VJHk*Q84I?F zXDZW9t!eOhYIuD-(%0(HvFsw4Nb;F!o|JzS<5s-rjE_V`Q~cKS{{*kh8ih$@dJ3B z)C~{GPQs|Y1Lbyue8ir()3e}dveOGF_A*qZZba2Jjfy}2bzje==jBmdsbry+<$>n7 zFuzNZAov&oBGk^@9Id9X%4-dNXATfGQRZ;EOVqps|+!L(r^7${J86KN1 zG1`eEv`KVbcp3&&A|Q*8F#T#Ck`t%7j*kxs(2;uhUKJ2~w7>Xl-6;JA5#wBgOY94< z(KCxzSmX70(jM=({j${r?tWl~=(WyoUq;*_cK46F`>%r~jB(8P&?RhhXGlmHlFhzf ziyr+G!($451-X}&T@QC06OX*{d33C!Xs187Ln+F2R-QKo9m0+3;pjl%)2?MN{OH#jSMzOtTwV%1&M5T&pZA1qU)p{c39J zVtbnYxQ?iRA#A91FaZ&l&Kn6M_#%Q-L;=rMFmJUT+Y(6D1fr-d{-hP9WMs*IsyAFF2ZlUqD5(V*ZwID0cGTl!GIW*!VmX%z>2hK zl0+DrR;E3t$HctuON7WCl1K=A#-jdJq;p&4mowGjGk?j|@eFy9ph>X0yA%Xu>IQY^ zoSf+b0=?ut7r;{Q>-+v_z8EK}?L;@7r5<`k`UZ>(wbHU6eXpde4Ey&~u2kaKTRA*= zs%L?l;;*5Q*sI*|fC7LIUOMrzi$g+e3ppWCD_}^842^YSxG-Fz(G-ik$?OOLlO}I|_x7!`5ZJ!2$)$B{Nvvez7RRj4TjmHh&w%QfFh;nK zY1m~l35*>O*a$ovsc$sv19)UGsKhM&zE@D&&#ck~2V@7K>&)vOI(_|mbDER$sN?>+ zOo-lSSx=Z#^}G3*i(+l-FJRd9D1(*yvl@j}%XCzDLR59JRpCPOM_8UtBo6qJQL zEJr+Q^@poTI7&Z$if64+-P&_d;cOuRyCWynj4%0DtAqpD324XOsQx1&;E8dfaph{| zoCys{2UaZb)TH|fveM*BOZp5-(oISYLB4`SldIy?vj*#VPKTtX=SOsnbKSp3xlyd* z#IZDN%rkQ_)_8Fv!0%4hoK>Wjw(L)79Plg7L$6~!V*qRLIrGxs;# zI3;cM<-?=j?*@Y(&G2Ut++<1vXujeP(aPfQ#mq8N*2mC-JBmDV73S|tYjUty@~s%5 z_JA=v`4BF^0b}06JxXf?((Msqmqg4dF_XSD(;!<^`8pSYX(5Fo%?e1|S{4Uh{AqpD zdLFQq_@u^8I0l2|V2RvbnL;5C(*Kx$Lx)qyo0+W}pbCHb+4^8g&HEg6^QvcQ#Rz?^ zz>~w^he2ZjExD<8R~M~iGNNbZ`At5*4KV4CH)+eLg?k+cgSGLd?{wrCfJkH zKSW-6qa>D7k1n*CxmCX)gi82*gr6rSre$a(U7Ngg(LC^~5#j5T>K%(?e&G^|M08oU zc5J)%<#6x*n47&5G3%pe;aIM#`$;hNy&oD?>jd0An*S4CxQDQguk=M6tS21^6D29j zW}!|cou?!RsQ$RPNt|+MhVf`_L0&!n@j$aD)kWo`^5&stta_-10as@FWRWlf7<6vr z&mm^GTf=pOoBC-^KcN_zM6RxLCGM(BsgW-n)I@T?rfu_1qz)c4LAVoe+~jlVpxaZG30m zCOQY-ngcm2w)c=j#hz(t<73%ae59~Y(f>y)fTYh50^s4POu66M!1rvi$X#d&K!#C0Mr)) z_U*f&N7Ju~U&KBo9cyn7K2Y=yQSoB={Y$Ft6AakKpRXoOd6_C6xi&`32-o)!pV#U9 zx?_L9_vK||#I0Aw9A+mKUQJqeT^u1n>&9YfYmmUc3P>EIUV#%q>V^kp5x@_`Yi?>~ zhb0R+@|`pVBsHIJVI;y>4|)N;^t%AqL}+w z`J`+~`h@YHBEfG+<&(+qsUi@d-`dTp?J>Gj_xFK8wAD=Tmmf#iutLusqSUUe6MtlL zgf4>1H60Qk@excg#cm@NL0+oGSX(mBfQBDnJ2*}BP*mjSlSf;r=Y+vj_HvhM>AA}^ zp2l+Tnd>SW^-ah0s(=dE|9=>JrzTOLrOUQ#+qP}nwr%XPZQHhO+qPY`%eL=69dV;i zcl5*euztaM%8_%9l;~vp(1Vl%va|uapH%|~;Fm?y_enT+U4&3A+OhHB`)8>V)Vplj`>Zev*JuOsdcEe_4%n^R+|ol4-E5ZgV1T5PuEE4 zDT><0!fQZAx6X8zBvah7-B0T}G3Err^iJ)ZPtGoCL*N7J%FmWc-!YTX@AvY9XcEAi zn477r3qkl-`G9pgem`C!Q=(7NA0spPv8NT<&_5T{KQ;mDHf#3lE{xd@VzlZDB z|08{2`acs%QbmvzLH|uSww#E8%tMn5wo~rP}jorou6_`)%CeSt6JpM0_#~r zeVKFav3+CT*=qPYKpFFDWxE#h98}R<>0GnY(xJ>B^N)my<60f5*xqqT^~*0&;8s$5 zUB|`!_+_a0u)hRty0e$>?&4B;RBpAbu(-|BI^V@hO|hFnXt{9JGj&?qlxV&Sy=ZLN zga{9)vpKuAtL|73EQZy%zQ}>Eyy;nHPD;huwXVztt8pKJ3DY+`w97$1-pCs7o%>c$ z?RieCwp>%Ib*ut|c&4&*X|r?lr}Gikc;z0O3PxOM#(f^(?k^Q(`CEsaS8{+*&y# ztZJ2OPmk2)qwSGveiA_$FLpH4ey_tc&wx%hw5Ai$VOUr_3AAt^x^dsJQXrw~*V|#y z8Bi6C$BfTgf6wAqH3|R5G>x;Qn6r98YfmT|2^+IU3-T+t~s#xz>9-D5Li4orr*>LdfVL(_M8vJ3w)s zOhLQN#&MBsIGQ*bgYp!br73L!g|8mKo7ZZe_O;;VUx)3+Iz9fub&lc#*5CglV{{xY znSzIBrvr)OZ-cFSz_yv7zmM_RV5-hK8+?6}R4T9xt3rSfl;+|j z8+sw7yBpT(cLqk|Kn|T6wnXqzDCJq~l<=pAEv61B{a~AFiQJwvZ_a(?te#et!xjc4 z`=WG!e`TeB$Q%R%CgaEe+X6(JtoasF_x$#(I=V=;NE`omm!Mq7l$Bc`nh?DKa?HEs z6C$|Y*4z~NAza6Fxf$g>gkXgv`DpejC*XFrp=jKzCoD~ahq!K%Yu0hG!VQHRF8uie ziqYjHOE~Qf&4U@>dE=2fvtS3;$l!H?kPd@g#(!mmz`>yp@-LjBI>439#y~gz@-3j8 zemDw1Pu=^QMX0$K@rQL5(>nma2}X!uRf0|c*+K?*m@fc;goh5NEIlUD4EK^NtZ#gt@8ji)z2A40 z@!;+|TRxTm0C`mOwug8Hqa#vu5PH=q+wv*B9y%LzuU9-TPdFX8g4CF>mVvqkJcb4D zyQA&cK8GRo6$V?%i~u6voJD(bGvB{9&*CI_HdCu&6a34+3?SMK4=J;{XMdX+uiY*G z@4qby7glt6jtue)KHpVp4%=m7V)X;-K4lIxsmdu{9|K>EjuPHF z*dYai5yV*zc?n={@JMDS3btGO_`>VYxzPJYgTV=H4gO&85)goTw~_lFgEn=8MW4Ff zCRm$0|GVlPCP$CZDF!ik;Mf2;z*jLzO%R|#BabGafTcqEz3f~+Lv#}mN9c;^eNVGX zNWtJDkV%8&w&9Ffns@oT+eEhoAV;3;2a4QdQ5j!Huw-Kc<55j$_=3ZN1vf6mOS{(q zi#>0vFHm3Q>JcJW{6NCT#KmkMbI-XD1_(&(;r&2UD>ySj}zXPjQJ9^JgE6oD1C0d#*j)Kmh1Vnoj1jA2e&YX6G^GRqK zlM6g3#78>mx42gIz9gs<|J^8(vWftN)Hxbp@bO~|^g9!$VooZ_W>uyvj8|$7*71A- zqp56>#jJ2{*0E?w9xBBZ?eRs#)nTBI$An(F^+mI4#QT(*v^2BhbU(GtI}D(RTKu4?F9oG!|SFaU~M0;g=|TFuaKiYjU~$fCWo(t1*5 zSPS)(Z&SIaL>VppT>Mxo7gdhPp%R@$&fU(?M3|(_ixfA?!H-a`ubNLv1|_GZ8*ndD z&eC#l@X8;CUcL~?4FR;nP_gEvcLVu^pL8M%z6pOF*?I~CMPtIr6JNwsIdUz zWnHq0*cSX+{2CQ(SRetP{yKGiT>uVRT?u`-yZ7wSMz)S_jy^RaUKu$iw%*VV`<}QH z2lwpTLn{Y$EB`Ux8S?~sVNQeOJw04tE}fp2U)YC6dRDrxFl$Eoa<|x|_j7(DQ#!8& zN7a7vQBi(vE_c&BpSc{S-O4_fuk~N$MUNl(0;7S$K1sEV1}c`qq`j~-#y!?UpS1|q z)Ej>Zih|DTahBh(VOwjs!s9}O@j^T!bXvL^u4*M&mP4h1nF8&r8|}E)I@)p22;XCb zC#Vr}d_@^6I2A+5hRyT z+y&0)a(grr@POIcn^L%fjD4kM$m8`~H=|5(BMq^cBLgaB9JDYTZ~ctKw<4ukLr&PD zQ${#H!>iAHe<_5Vqr`MDDmfn;6ju?&1`zW4L+s8B?ze@RF z4Kk`iF<|UpBcFk1yccmM|31*e%7|8iv4LHr zW~gPyhsH~!>BDvLpvE2{O6n$nf&zXO{q&IIq`|JT#kk;bYnEeD8OM4)oNbr?7(-5< ze+_VBHQb%BDtJ>2X$dz{4#;8jXPoN6JB~#wHvR&bX0bYk~d6vyOtt zeTH*;_9XxqHkjsbjUmPkQ6xm>)C^LU#HY<-jsLEbK3Yqq;-^iz{A9)B3BK_VN_T%< z74ZA-fQ%Se(_+aGabRamgM;Dn@( z3#S7zRHFhoQwy-&ta+y=Z#yw&j^47QpsJmb#w@0ARdb_eB1 z3_rY=oeBJsHvDoQSq8N8eT6GRC2G94IlnIl5zVA?G}fhJ;G-XNpor)fkM{W#-rzj5 zXk5TasAF5{0X#}77r54QI?lbW`wD)_*{sGZskKIdW$W#vKc||1 z)BF4|`t8P?i*g$8FS@Rybkb7|QyE>%Bn)6G5)vLw>Ip4r`Y0p*;Kg_*il>~TFtVfW^w%X44EvP ztq6Nx+r*)0V(3o|E9LeO64+@5Tk7?^H~=7r`0f#<|5aT#%=Z2QwB3aI$NIXub4wmo zY7ZPUgozOtSK}$Sm^r9TH3@1M(suIW&txw%hJAdc*3Lc4jwt;8?%lNQsOL!B34~D- z3v0x|U;xKfGT+r{Xj|xH(Zd_9@6`8-jdTmL=da3dZ(*A{*>)uPB_YY@)?I0Q$ETo;X9Xz zV~pjEaLF5`f=tI!nczqwUH8h2MWjX_PalPG2w$NgYGd-DF=j>Q56@{?@SiqPa>_h$ ztI6FFugZ-98@1AgxFUW!(afgFXaq@EXR)L^8vM4>mbKYfI?FVtEdozPX z&AUlnzSJ-sDXL7o?#yw0{pPjBOE3Dh(zOQ9^u;6Zmq*BIIQ^R%oox@)DSc$%V2_gF zvEROr$G(3k1{@-@3&lX-U+&>Q0Oo(AnT!9j(+)BIckbc8eUs<^%(Vai`@hVz|2!X# z+hU2mpOjW8j0?HxV5DATfAbYAf(j|)h&fnfBX@^J@hSr@ZW8Ay+Mz5GGw+nAp{Es} z4CQ;TI*j+M`=)6CK=V%$#1M(HVm>A&^XU7g2lx6E-d>NJDO8S@Lv%f^@sq2TV|M@P z^7!awTUkxNk2^l<$kAuR@-{tm_{X)LMi z+1OZq?j+(^d+E_PjcTDnLt(Mp(-(zZR81$}p zWE{MUTbYngmPc=$L02%9%(N>a*k0idg+pd{=~ZuBxrKwEutkz8|EP{oInnY`6sQC( z3B(MXy@>3nDLzI74k1mo2F&(^RSIdZ?cNZ)jPK5F#nj~Il`z3(Lm;1d^ua|zTw=y%$bu_HDMJBx3;*!DydpJ_fa6qI6*y)I`uiv4lt-zo zD25d{ymLcU$zjRZK9+_62e00+a!xos7y7;vxrKOm7~i`wZiw=#mR%vf-LW@*D!ZJ! zR$6rJ=<%FP(C4nwj|qi`qTMxM0U{HO2^kjPJTj{g`>E_-LV3tMkaCSi9 z+M95`g1YORe`ZxibR=Z7Cz|qqma2Pu+@<*I>>xW{OmqoIQLem$0W~776;DteWmO8z z-NP^dB}m4o(89(h-(pC~pQ*w)TS!&n#&99mjD-qh2`~&W`I_P~2%;3joo7lNk_-f>jisq$kZ~OPDc^ z$O;s~qLyY=LwVG&ysKbk)8GFs3*XEc!M#XII6_NEXy>b1wJtjIe2nyomZGA1t@K@(O82JQF8?sYIf`1BCKz?-7+4`n zeISA5>p5tizh9Dax~)vAn%+rYxFT!}xy8K(dWl{q>Dd!uJ4vD`<@iLdh1Be*z@MMG`Yb^hLCHdZd@yo}d3e_F zsuR{jI}=cr+{TgaD#j8E8X^zU%f~%B075eE!6$ZI_EJ=35f%j6s_uk6gi>UT4IpeP zNuB(n^AWn5{UxHO3GutFb2Gde2O`-xb&QRkF&6V9*QtddR^Egx7kn51DY;wE(b)~0 zz3nv*XBQbhk!N0!IwFJWKB#YVMosq|VcA-H&@r!5UY#6)ug2SYgH%0}=^l$57z=?5 z%aM<)!^Ams&hYZqZjO$h7!YMt-5-TPcBg+R!m8i11_LFG3L6;tZ4> z9C1*dAihow_#yHjlyL&o6|%e7Rf+_w_BDGt?0zg|LU2hXxM>7KvWe51wu-5it4=$& z3vf#vKdlINZl_9T0D|f#CgY?ty~k+(#;X6FAdb|;N#CDN15sQJ7wPOTpEObLNpm{g z-lnV)$HCVw4%v136f;+3!LBAsN$=gj5C61|R%20V1_bKp+BHLcmhY2|Ci9?7a6M5; zY0R z^S(AvROx@VUlS23+wovYB~7#|S7?OwCemLO&T~eff^gJ#{{%>Y=4@0bU@7}mN)QZ$ z>0`joDEuP%6&o=w3pI1W#&7#Nw5#Mnw;C9fNyVfz zkwFl*3BTYY5qeueaGnpj;8h6{EQ1Mx!mK1#o;QeYq_B(7vd3G^oD*7zZEb^zm1^Jr zz9L>2-%8xd#v0PD8TVn&HjeyQS2o;z1H&{5F?fqr1~WNvpaO+y1xp4QqHHrp5S}2) zub-ttEeV2k0X8f4y(QKw=_d8Ysf-tj6xq?$r8NXGB11}dq*LWv1#3%iR5*tHmc$ad zFC67=f-CiN=OWSa6C++=C&jZzIu7DP5^d0`F%Q@jvtqtu){OQ6hQlXA%K;F)#dRFN|-zxLr<3jD8s)}<7Ell zX*$dR+>oJRg{jwjQYx%Iu10a$Kq%O;v-GNV7z(6prWFMOJ2iFzy?lIvBvMU(dD*SO z0QXvtDIw~30t@9@NdQjlsBbfi7R&HP=lf=5t4D~P>AHGzJ@OpVD<^Lf_9pt2Z})K$ zq>l)8=pDR*;oeW}U&IttHV}>_Bn6GRTTk&JR~;roy>icD5?8P}njNgxPuB>E#d z8MkjBneoiSLJMDn>|tx+r0uGSM#1@$4~)8|1_CA^T@@?~1czJ|==P zC3rhI0wtGEw`oK(c)l?=E0g@@@PGf#R+_%zn8Oz%PfN*FloDuym(1MA6Z}CxX z&DGl!N4R0@Y7P&yZqA-^qzX_Oz`Yf+JFPt{?d$9+$hT^uD^m-Td5}e2NEqu&3oid9 zG9Oi6$^=2%APQ z>zaS9>B0R?9!b9XC@WijXO2g1ojElgKjS(QvRQk-j$Y;6FWf-^z~wl#i~bOfV|~X0`g^!O7G=jpy_6pnXF(A} zy%Mr8>-NIEg4MM$-Y z7r)kBJo}3;Xd#gnkAkk#Z;nf%)T11(7pyc3u$#>6sAeT!3$l+o=%P9DEtZ|X{md(M z6wSC|!mJOk)#xa>O=8_Lf~xOZ;hCeVyZUmoPWC$auGs|{I~aw-cSv4mfFUP3DCsa* z6fKf@VIPQ{xVLg+ZL!Js($CV8RO1t|JIzR_*j#rffXYlcb2s@+P1cMH*F~%fKi(US>1h3tcMK`t2IpIpJn<+0TJwHkJJ-%GRCxOrXiN7MwW){M9 z`9^XbRVPWPaP)Y@2Nro+{}q;g6~|OQ2h~Nw5&SkW(ni)ZVND4SchftRU96MZ-7WBv z27>MV^2lA!Zgeb-B6p zWegq>w9G98M4R=XdDDTL(+-n>A|EI`|dwB>(TCF z*Cc!uIS6k+VNXGB&E9?^xx8xc*l4ddjVm@IXZOwJKMyRV zI(Is}d^&cui2EU{z%Lb>$KYYH?o4D&p)P+6CwZ)J8|^M?$Wyj~t5Vs%JbuSrZ9`Q= z7K?j$i@BX(*lYC|Y4Xxx{oavN;xFpoBT1uavd(h2=h5SXc7_$s(z!R)&>`=M%D@u7 zB55O?V78i@M-E2S|8m@;^oT9S&el)6i4`jlz{T6iSP4CS$YMLXy=?W0pS5UjpB1t! z2|ug6b{<0Q;Y(Vuj^i}9;6IILH?zKNb|H_{MP zzT!A-D7uVmc6JMQo9p@LAv_38ISZMI*n+h;R(j0PAIBNB-w$H>$=n2jDYI2CDMh>B zI%a9MPkVKC!H|g2yLjxC%kSx$t~_@n*#bfyjdd8C*}soAZX+RHQPwWOb*4*WWUfmZ zj8#uYqGxTi9bYRvDxvQn4e~8q3&l!3r$KuAPd=~8^+iu?T(thyE)W@*I0HCebIGg? z2xOibF(SW{fyhj?xgrd3k1KIj?fa{ZP`N)jO;f;GlBtsG?-Wz>TmxAGheC>i;hb8{`%U)5$$hF56h4Ofaf@zr;(Le}YQMJ< z$>{nf@zLw;AV5|L-9aMk*-=R+Gn#z-MKK@_b+|9Jn-!09l~ry89k;ZatxD>Gang_4 z_qB8uD*GCOy}x z_T1(femo?hI34PI4^9^09$6h8yG8o1zXdm9q;_KRO}WlGcVQDtgf&y9f1f)3yfOimmJ&(PWZond zx^6oAiHG%E5o7MuQRNox4uK>Kwob`F;lQ0#C67X#zleWFUbs&INXG51H#`r@PI~79|skI$l#5|O*WtOUh z4RkQ<9eK`(mDh~Qq|)9juv-}Q&gu3vntFLYKSh)O)nD{#bdOZ+@#Q3J+u&PCAgPXe z&L?L%zh-QfjDI8HmdbmantK*)wKdV`HUi|(Qk;f=JdL|nKfGFo7~902@^#cNMqf&i zCt!b?ORHRrvG|oiW5?x-q_AqZTbPG}IuzjL)Or6giM5bx!5+1kd+?=e1As1F{mNfg z(K5}xq8t1xHr1VwO;{SNa2<2$t#m5Vy{sF9HqZ&rg7HLDJU6m1NY8_3gU1;C1d~=x z3(20kp@i$jNoN)uG@&y|Fg|?DMz)#!S^i13(@GNhfxoAPwrdYeEn~j}$2sR=lh*J= zW{FqMwXUl?@o)3+IgO;MxqF;10$fPz#fddnY<3Q{{oPLUAUSmHEtYgEQ;LG*Ifz?V zx*_y$^PhI(-we)k%a1Q3kK3q!M?PB@S|ZGR3V(!{_!apLr-1x`U-85%^7s!>=kl8A zm)~P9KR+MymD5Kx+}nbZZYYI%mF@K+k4@Rvu`WWiD{g;_>)%0}WmBtzB~3lgT$q}w z@bh03f9vdPYnxhvAP#P`TjE&ksRrijMR%_Vd9Xd7j!ZPP{3d3RE3R)T@i(PG8K-Ab-3L z&231Xc{;w2)|WNoS_UP+0@VRHfxw6n!0sOD$o35tP`}i9X704>hsJ2qn-b0OaK5ogf&Bd!h;8@EBWlL_bdQI&$k2JSb7cDO6|`f z^LVj4&ukMC1hMD!d_Y<}gdF^{hrs7eEpDO;&rP0hF|s@Uz`ZZ5q|l1W8bV4U|4Sw) z)>WI271A(mRI`A~4c%4m{d{lrktz)_MFPtO%v0Ec9bIdj#95d%4?^APDfR<(Nc>G1 z$XI@rZ{p|sj+1uD!Gq)qhG(Dxq;+LXOCZ$StaX1K(ghPzCY`)O2#+KEuB!+hPl47| zOBM!PU1ikUq*Y@pHRyIQxyr8x_{wa=QIilTcPPN&bNE!;S2tr#F0=KcCQ1Zcv%4F| zr}f9X$4g!+w;$m#`U6)FUQ`N2Nm6J~4$H7+Edndaxypa1*rZqFOB}bJ{?F9hCto65 zxP}+&LbCG?uA5Cgxvb~IAJ7m?Q6Xr6D@l1;xPa*@*{Eyj`?+SG4eGy#TJX}Ta$4tN zNF=V+HTp^eA7ec2Wm&{}ff`NmJhMKscb-2jr{bloiF$R&FzqOO7Nc&IUu++j=seK=F8( zW(`DVyx;Jqqrj@5z|hy_E!mDWos)@mTrnlk4sM-}R#9T&1%Lv2FPF1*hGi)b_6wSuv@%;N(`qN%a`yZTJa8q{jI9*~H(mQYHg_UF zzt4xW(mKXffl3@z&MYC3uk6U{{`9Q0FMl2!N4_cyC3Y(ZQwkc z6Ns7^6w$jNMsw3+apNbG5>4hRk4@ss2ypvo&88~B#YjJU8A-g(vi)6kI3|d0c25q3 zcq8KFowN(t_?#ZGP->NxL@kaeh@IO~CtCs)h^{X@!A>9zpMu(vYmEU~o!X>sk(5h} zQHMf_38)zAKpw>k+6C2lmdBi+q4GGD;v6d}`#qm&5q%C{Oq#UJPTL zv}jQ7Wc~b4fV&*I-}shdlgsu3xaXAUQ3ylX1?oXM&-NQ<&877Ef*Ok$P6beNpK(ry zaG@DbQR-5#pFnGRNRr}=(<=12Yv{*}PH;2J%2WBPs8e}0dLk^RUeQiU_iAKtv?xXz zRaps8Z=*_TiW-v{1ZnKe)7jR!^xMzO79pS-XL3(CE`Qe-6)j+1;Y)NeHg2fLbftEE zAx_BQ+i#{3K21*oM}Qr~Et%nFIH~S$4PH1?KNM4@b*NE#s69g_Dnq zVI`OgYj23R?jOUt))E*;e_~w850dowc2--umyVn7oVgB;l7LY@&;}#TV~T)I%eLmLf}t0-38yoC^>$f2f0f+E zGO_wsbJbw6uAhCZ3~nDrym*>{@X$aj3V7BeMQE_*t9zlTxyhkw+3baWyR zL_3L${WWL~_mU}y18=-n#J`qk+6woUZE*&-6{YaiG0CRZC!<{KW$?9|6%IsrzZlkl zuf2yt0$E&B$Q*N&-%0w0%oA~I(4%TPK&5O1qkHMq1TzwiG`ZCs<4|7va&F@&F1+!# zZNzslpyFCeb}4zyPq>h?kg!z5OPSd)0r*jj3*kBju`du3b}<3>NKfnZ%MIbuWy*Z6zvta z^W`BA`VD%PbpRe&;NsB4wsdD~(Nn>0oo52arosJ0Fqf}gTDD&+E77KEDMtQq#Ht4r z-P&Aj(w7Ll?jIkfZIVn^$+MR& z!&g3Eh-`ngS)eo>P7#I|$(!Tb8y53`ngS>!*2)7}%<0hQG`UW=fP1F-@)0#~iDJ1r z`E1Xx@3hgmwWaI*rSg^UPAp4vvpWM4H(8CN*r$4;S@$8)h6OzIE103`1chFcsI3hH z8^e;+*Z-C^6H)iedXS`7&8|=XYk%%f(R_C|^qVQQSbJmN_^3PRbrt z-@HZMG3i=|W$VZY(8JsS`o0ZoMX-Z;W-j9GAkb`>^6a5w%Lx0a2T6KVo04@edD=<0 zb6@&(q7ECg@`e%gT&|{YBcOgXIo}UdOX+<>_qa58FT&j~c&D7sGF?(rr?}d%nmgWK zB4YN4jh;&DrKTrLjFbClhx8C^H-@syi=?qIL)Ik@L&vA5BZt-q_oX}B_XKXPYw}IZ zGvd;T!9x~}B?I^Trgmz4RAP5%NoiYcjl(Vbm>QxOp#Q}VtG)6#_|CTTE+QcKG749= zehIpH2Dp#PC5e#%iPE@Gyh#FHBX5!M z=!DwEDVR4=`CtHegX*YOqx}*%39|#orCa!-U?MVp@Ck^TyNPHjrMwk)XrJ> z>5`B5SHk0xcBM_+M25l8ClGJC%Tm`q&CatP)|*5dUY4(e>Lf+ZH&fY58`(dT!Pvld zK%-uid5xmSqZLjabkR~d@~9c}tGtXIVZ|L9p>C*Hmt!>h?kBeW@Hmb-g(}m6|&v@&RkUdlVmyThCIo_JO^!e5*Ih{8jT&oRaYWr z)-pAv=05L3SC{<^B$ID$%m|?F{>4SQ!Q_~upQ{WbOBWc!07x_GN09NrTyIZ0ta7BAOqljG2f1|rMO2q@&} z6RC61SYNpSOR|2Iq_M2;_fY%jU1e5Tt`s?(U%$sW`=9@Hc(JW?Is@?^YxxgR`G2z( zU)uj>E&s!VmnmAZaX6fgJ#%&Wpcp;SZjzgtL*Rn(idH_+(Yb2!Z6x^7y!A{Qs@W4V zrQ_2VcC>%@{fs=|qc9ERfc3p=`2V6Kk|b@YelUQ|JDVm7GV$qkL;Th6Q@q%1{8x!u zuK1U`u}`^NF4@f#HT79Z9~ou&?h(J>zxigksCAWscdfGasn$_YZ{5VWyjQnfzB$s+ zE*reD$57dt?37z`Z~3HOX_xv}blUyS(OcE3xvFfiV@OdOXzpRZ__jzU;|v-o|78>e zsS&@nra@N$ulV`4XuJ98>BE`o^}`v;uhn5MH{idzb>Dy)6>m^)ZBo4JbfC7qpRW*X z>)g6am55gCdC@xP>upG}VXqr4i;~*ZzG&gv(Ae8LPjqTcJ3!A(;lTtJ*Y1o)N zS4K03p}vO7cADZ^$+}Zcw+KX*i+mfaN5=R&VPrv%P8L1We;U$}dI-?mztr82Xc^dp zbeM8EG7+1)0LpvrgeA)zeu+Z~%ilggsF#KnA--Ba4NgqAosd;4F73)>cz49$`7ObF zlgCs0r=7gahCTbCWNqv01b(@($ggkj8|$u*c@2weFsgm_2%P8fKIkXHGY7>DL^(;m ze=p7ecsN)+<$j&jfZ@KNanr+sa$@mp7A<~U5N;KNEA?31l z^oW3&-Js760!U(P8NF-@8Dkck!kF4HZh%!M+=rG+)+rjG^I)dtN}#p6xm~GhBJAQm z=%k~^jlFe*O*<*5niV)I(`f$Wd)m0xDQE;ceceT0qy#p9>Xy&ZqBsK46(Qn|GYI|m zsMzkcp#d;n;7mCS4a%m$ZOWa^eO()7F9Es#4z003s=VH6bjN<)eDhpc_uHh?6CXWL zYoaQ@cIM*b9Qfn)ixwPol?hfeMxl@`I5?EWu<+)LaTUsV%Fv$cXbr7@o@&QHnJ}kp zeMG@_`Gkm?h?Nm5$zb5krQ5 zK{m*b;W(ZSq+@hIiLE?oZ$NYZ?a5rkf=1icd8&P`-4!(JXH%8KyB#~r1YYZguRgM=1A5-cuF@ish z!t=?En?sgBx}2q&6A!ErkCHkR-6VCFe!a&TyrIAm#nJ#ap*~v@OCSm?GJB-rdwRpn zHUS0J*W1g%;T-%gqC$^fWULO*0wg!F=u2&p$_j#Ezrq{S{N9@R+@48|$C@8gf*chTd zBRZ#pKrK@WYI$LDMk=x8xk$0edUHw91k@5|EsKzxA{T$U4OA*RBa9NKxm-W%X=imK z7a)KWJpdnQS39yWBS8OBE0;O=PnLldZUxBj?#|x7Ley5xYm1Lz`k9zWPPDXk5)A&b z8vE?y?Iv>&w-a#tbjZP3Q7}*2YbufV4d3fZ1+WI7r?yUR=V`#Ealqgq zz$+uS5=VfnF(R!fnJ4Sg*bk?;*vak)hBey8 zc%Y*j;48Y#{C6bL!(Y_FdpJp-QSFbE#Y!^YK9Vv+JvU))VP8!G2{8^#16)bcS)_ZV z=F+)NAc$lP43s@EJ6h=rC_B)dUfkCw{vsxuM*0y*Yp-gIUWQp77>*QUPX*ZR{(12T zESp_0`O_9ARt$w&;$`~qBsz- zw_OlQQKB0!$83)cYGqe^FX^y~24OO0mF%J~V|Y@eakJ3#P`~q$l&%#W;!Fr){Nr=FBJ+3%q3vuzwv0 z#ubj~D8?Hmvor8mm_w^a=kc=`6FV>kRJ&`k zcfY7-EMphZ6Z4pkIY&eWTx0cf8${yz60Rqyre=Gg)gwRtza0bRz#u1dtey!nI4}h`KN*bREW96Mw+;#cuOhm z7AJvvB6y~3^rk;Nc1tDFVCH2PU24h5a^mTnd*Jcv3buZ2Cuk$0kKr(D*3&~N)ShUd z`kmG}(U8UvouNN?FsY%3^E71BXD8mZH?_Jx-+W0<#f~@xOB*sqdsZLY z#*RNFxw_%II4Asv+qz(suu3pc|S!yVTFm^&me!&%t0Bv6<+ zB4blZ!?|_vW+sQ5%iL-O5hy#>^DQ>}7W3J0jG0*0>KhRL6n7DXn4xRGz+`ow^`bj_ zeyVtxh!P)`4Dj{nSpMoZxYn50(UsrT30L!06y55k$7m2aZ4-G#`Jc%)Ay%ZHApU=h zy;GN9QMPp(wwYnuwr$(CZQHhO+l~y|wrze{_vy5D>!E(co@>p%#~Pi?Nn|7d&P-GL zn5ULIoA>4kem^do{B3k^=e{RtfbI9`1J^V~3GsduJt<)nJf=dy*MCHob{7&jK5!w8 z4PFp~K_9bsuRd>9ybXCxPBAoJh~pOM*ZI-Ijy#H~UHjxF3?|7#@sRdkvluIrZZa`u z^fW0dcb+o;WGK7Vk0Oz38JI?e(6WmJrxR^7z8PH(gXm98vR zA^Ck!^$VQ-3lam&_*8Ddqx^=fm_<7vuSc$;#%Q4^q{B$EO!XL>T!XKKK{$ga?fWkm z>=jIT*IY>*A&SHhVrkN8B?+KFw&;N9Fc9)(i1r((66JZ+tGdF6GZ!k!Ymt~rHqTFB z7D7@GO+}N=xWw4r_ZYqe&enRian-SL6?=qEc)mUFCn{2BnKSQW!}^rT;?!xkL|V9m z#a9C(J|=0 z1?*?T_aJBWR6gUsgI$WGEhGvN)R4Nf*ZqDn1lbPHEKZzvV!_wga*JyXW&C=?fwN7y zia`R3FYF|1V(XsC^=O@`Bo8?EUoJL6Rz=-rqyU_u+`)|ZKqI-fXsCNiQ}2Ta+SB`) z?wi;($Z1XIFZ8nN(3O1;NyyZh?Vd? zIBW~3>YP!Ht0;b&_Zfz!?$ zk#Mc%oX^9@ikt;J}8}&wDZQ!Zl)DZiu5()*rAvpe z%g1#MQd_z`l*3hQit7SgfHHqs3!$E#d?@ID69CAD| z%K4j$Hi214P&VuM;B_H@xQrhbxzvK9-E~>p!uFuqi}J{0Y#RJu?$z0(-L4KeF$TFJ+%FQ;;UUTD-c#~8Ug+J!+DRAh`jpSzsY?=zz{4(?D z`U?GL+gLt|nh6UbKBEE<$+-ibFTg1P^psXzB8!mq_A;|~qf=L&isXGisv!zvr5w*- z)fAJbe>mn^n)(Rzy9-u`?Vvr6n|cmiQLO=(?wQ(zWX37FD6H!?@n4Qz{Og10WaOMH zrCarM_uQ0O#9KF4M|wt9Pw{R-WAG~rvQpNUR)Roa{#qD$qAUTvG477TQd9&_!fJ=n z35pHMc)pE`99;7qx1#%7>hKUr;|goDi<=W}rdBuzjU$DLDzh(3u}qtMdN+1k4(A|G zZ<2-7#A5JiV|yJwoNvTA81ailk7)~o1f@%VZTAcNy{VRO{f6VjuB7rw3M6;uJzr2< z(c6QfgYcW1Zd~1AAYQj`eA%zJU!smsYF3svPu-iZ^VQaFn!_fxC{?S}n1yRa@9roo zukW4jKcN;KS;=6yf1u?*IOV^C7FEIj3tAjq{wMWMjkb*)mYVBNt{wxbJh{Ng+F!J9 z{XewWZarH4&z-8-zT_VU!m-5`6B>4?-U6o{y$q~&pqoe)_J(_XN!p6T%*7|g^an^)S$vHnRs(b-vIEk;TOP#pQ5b= zVrHAoKLiZ(vt*!=-A@GbSl;b-c2Xb8zdP5)1C|RDfwEtCsHg9wepz|s7sI0bU3gTv zab{l4W3+zb-qwq%@Ml#-hK?xyz}ypkpiN}dY3pgiMYCnRrn$23Oz})#T-7dE4GC@$ zby$}U8?j9_{;vIpO}G-lXE=8Q5f@q??)=)%re1))9*mm~&Hw_qi+sS(U>hc`xqsQNNm<#Lto)P*!vQH7%QCbZI$H~M9mob(KX}NtP{RD5iYf4l4ABe zHTB}oi@FrbHm5jcQPh_<9Jh3$QfDYkThwtl(15eQVVU2_un?-Xqa=^*zyr*&rE;5a zj>H1oae3HGN$wKrK-4x@Ygdv)ooRl98iy@LT+|Hk}!%`%4(F_bOCB(k56ICn3T@MzOT}d0tdNq*=>+eRn2VCB%%O=C`PYoG{ zTajpVW3CQS%W9&JlU0jAGI-mXuyew_)c46!StmSQ`(K9+K~jfTz^DuhAq04Y7w4_v z7skefZ=`_)zrFd(0Mx>15R7$)E`(tVc(VTxyz}F6oB?a4M63F`ULyT=JIQ%}@pBu8#kMhdNqX3JDa2V)yFP4Zn@rBeURnNbn9 z!IqNMxFp_hyUIK|>PlRQ1r$w2-Kb{v8a*B8B|UZkNcEQaT^lu@NB(El)h8G49du}O z$nMoOwnyOOF+5}&&XykD`2=hA!>?sWg7{Z&5w?VkN~!XLshT~6sIj0KgQZY~3z}D3 zreMZnq59{EO5W(#$Y~3mk0*e44aWMV#J(4p2v^)*;bL?M*=FS3Tp~JQ`AeP}Cy|d) z%v8hC{iOM2Niy$mcgXHl18%$v_>Mv2Srqjf8Pq#^goyMD>|;J*>%`=yVV6&TtP-8S zl}EqxXFQv__JdIDp8;OQNWo4YbBkd!nkx~}2eUU*_!;1w2DWE*{3u`olsIkaNx~Zn zWZcKLurH&)?0)U1NrpK>^(JYPBw%8&l0Hc2ZbdJ`dIrOL1OYg~1ais*3u-cM)V3LT zO8cX+G%LQ2uT|%>G|&VpF2`iXggHBXOq)u8d1dKD_YY!OtTw-o3L_YcgGTz)#>Ledm;dUUcEAzcJ zPwrRwK1YPI;dt$4gxOf85*rjb7WcAi>eMLtcT;ga#x;@3{`lZj_Q|M23^38a#Md?r z8V;@$gU&ujD*AX}QEWq^%6|vNSkgdkW@pR2zmV`=5%L3E@hA$*yG-D!F!rPKK0i>h)Qmg`EBH2Lmst|CrJnt&wI#l5ljaFnKZ#7;6&Vld(kmE~k zLJ)PS};{7SrM8pMjNEa0}G zaby!?Uw%t3c~-YL<4CI}^{$ZXP%RK_)qkPKJrR-0I~X%kB2aD5I7I3sWV+$s^wqMJ zyGnv{1C@}DSB=M@4p+5)qG{$Q0z5x!a0bIBAj_)JC;B-cq?MP&KtA39LT9(gtQ%vDS)mWC41@Bq zY;!c+vg(%`pg~3{>iRJR34hMW~j$S-1`b`#Jd|o=?L;IGyf6K>))7A0Ky@Rx?CW9 zzHi|+PVGZAcZ`GKDCwm^8G*3e{zMz9h!YKsk5#UzEcBV6&93i%6fx!aH@wlKRu47GqRq#L8pi~5`8;V;!iPv#3kkvi3(^-!K$zYoGRRjh zel&+z7U-8pAG`?F{s0Z6)~0CwxMhn6Himk(75&LAjZW_+*{gC30Imbr_sxTffnz5m zYR;_k#R=h~F#bT4CX8Tts!AL$WWzvKoFEspIs-TSk_PVgTy!m(3%wJ!Qt}l!hB~E) z@|e~_nOp|RnGIe=?-9hNpx|ECdtQt#A}>#!_O^KbJN(wCl~bb`(K%CLsn;Bd793BQ zl_GU(C;ehkb2=Y%Y@~U@KIIKlLdS@I^4)RCk~}RAbD^?MUprtzhypU&KOqoUUn4~< znaG}MTafe^n046wxY4%ZI#TUCw2+l_8F*QiEF-7zoGpY13PyhjRGHol3`n^(Wwyef zQrUF&cVfaIr(H{?NYAxM)rZPFluDrqXm3`nV34RxQ5lj}iR1!u)3u=N>J3I+VETG= zy!WOAb_<{#yRK!BbT?p2cAO%xOWPtRV?F42(=E-}t9I&55fWL>rh_^{)akgDC>2OM zXQ3k;6DV0^z4FtYSY;>xfC}q`XASd+PTKxZP+c+aXv>j8c@NH|cadC-cmCCOQ~WzI zo;+tx_BUp@@Dt)gb3h=+C=HW(0sq|*J>9U2P`C=lRms=Y<#=QHEwe%@v{PBg>Thd% zY^`o&olZ})V~jV%^Siniye5-qtI#>8g98Z~1I^zsK6}RqBFdR;>ct<5m^z|>K>WNYq8_f(vh9T{;@&~foB*xIcK4c+3ax>`(uo^ zszINL`j43vD}1o=qQYcEF^1r1a5UQZ`L8`)1bmZn02GDefa4#l^T7#|_vnv- zo)Rrvv1j&+ZYMWACheOWCc;O6!`N)k-Q>Rull3KKH};3ZmBlGqWHH$l){umvma?b2 z&}s6cw=SS(e(P-#W_fg#3^0QIA!LFJ_*!v+)8rrA;;Q$bswkyVQlQpjBs*d6Gdt&fzzc&f3vb&nzt zp_G)GYRfV)7Ws7{Ym2}gc?YS;s{>&#xPMGIGToIr)|CR;1^Ns(TM(D?cyBS0nUGJ> z*!a4f>O`R!)hfDMdg*DNEtI=keAQ6NscZ>&E5C;;k8w-fwk2UXowQ;f8M=WdPtEzRaPBskymI4`J^(pW5^K>ck%hP_jfAg{QR{~j)Q@(+<-QBX<@6dhp@-=)c?szZPLl=KV$XtF`Ti@}+(bEB4 zD+pu@B7*mNx><-@;hMVyT*kSiB@#_nc1A?Xw0sibl4#N#7g>0BmSVEGsfkb>k@HcZ z-))okJo^k@_(HSo-b+%h4soG8!TvdzVipD|^+0-WZ(vNT=2|Sz(0a+JstdDNX;2u; z5yOH)wFm(~XD^j>v?8@I4#k%lUw2}AZmnP$q}CMCxCrL>EnSbXTY%t5PStcc{KYdg z;EDQMDL5O~msmiITQHyDaaA(79?PTNI8VB^NPrZhnOGY_Q@M9SISlC|Y&z2%K*R^? zND1}<1AJ}UiN?Wj2yPdwK&J)_xH9eT<%UOv7*0qj(%cxNg1DmUy4}Vo;{0!2cailu zRe0T&uJsai43Wcgl;omi^zK>r2CqzC%;qeZ!#ib`ZN#lJ9T6&4NFAO39zTgS8vD1s z6=FU%3?e1lmVrMrV1Jwtqpu6jjK)uM=@`UY4CIb1GTk!EZ=;adj&1FR_({zh{lXNY zAGL2nt~xho0H0mCl*hzV-fv{{@xA-v+SCvmY}8x%MPb*+Q6fHHGt6CuN6|j26Tz)& zN{~fe;_v{D!j1q#i*!IWV4I*o^2u~opeAPCD)^4Dk5%iQ*K!PKL`D`|n8a?~1=}r| zB<^{ViEefO%nBZf>1avh@(ep{Mdf){mUWI}z)ULcoqrV6d2w9KfasuBIF;A6EE>lU)r=T_+!$ZUb!151M;c|roLU+b6{r-G zQQ{9Dbc$}Rhd;~K6~|1vm>RGE?`?=KKxDOeGR^lsu|%SQHb>G2S&g{~VBp@Zt~R$; zGaIiWLR=Z*l5Tv-XA6crHGkfQb=F9Fi5ga0#{)< z^b}U6z!NG&XscXO;nB$CZ1I>#7ZM^h<4g&a*RWHK>v1n;v&Tp$X`%hHAovaHQEYGA zqpr?sP^A8pWg9z@M)8~-bj{}JZOc8bBhq1QC?q_d;yK1pX_3+vQMr_Fk!G8NB3pp1 zt1RpI4L+tjWlMv?sZh$;Wp+{l9PNaD(rQNsIc5^bU+m6ypN4Q#ycbGy+^T7N0{mV& zoi6B3rt033j^U)?VM#xb%RTc_TBhZqu+h9+s-PTQsT)`SMP3BCVV2N%ZT#>vF&+Bs zT(*fc`J%FkaSm?V$@Vn1)Rr7N>%@sqv z?$Ip6kP4tXSxE9iv-?HGZK}KPvQmE;+wuViuE<1;{Dq0!&-~m0&Hb*tLy}vjLQa<~ zcUJ*Imx(3t7+sNNEou(^@C`Yl1^aHfDB^I!Kh@21#NJ#biRh4WX**zhA1;bwnz`Ae zxY=~SS+p|PQZ#h#)VYK3j~8sK!e2@4eF2NGtGcO$HYxmM*6VqgzLX&Y{7eg!Jb-s7 z4bvs-(cj2wZ+xccu;{pjv3yHSNZ@g+k@yN_y#YMqS*^3sz8{ehn?Wf%K_%YhKcpt{ z4r8HnJnbR0Tc+zG?y!~$xnz4-k>-6`VNRT05&a5jAF4J)m~$3cBF{Gq0)eB5(OeX| z%KbxrgiC2H?3u%5bXf=0m27}QDW5P1obyQOby^cO$Q7mR>@FbfUk3Z>B}VQ)@kU8PSc)K{pT zp&-nEiGJ0yS$>~X@u}>uvDR?Ahvl#=c?`oC?Wepq+qRJIpFnb_cNLBlg9}-6V4#@w zylwu|)3i@dP9%2go#hZE-PgKdvm9g8TbQ&DG};sN!jd?GrNCq>Qpr*C$F$u$Q1F#J z?+x>aZC-U*If7XlLm05(Z!XiPW{yWt>#0oQ6xMzYiY^faK9BFIXi+xP?$Oas` zQxjC3HM3eM4JwSoGQB>F7McU@JnMU(WJICJTs{HiES2t}l0S|kx95&I^-CJ7Oi~cF z$B3YAe4n@uDmiRr4-R?1JB1i}cc_c+%a{+*TQH^`RNb{h*!cnyI=^`U$w40;**OXw zUGN8$fOKE6%0)Rb;u|#rcRO{vpZ^~P9j;2BsX_q&_z&y+?-W#m`u|c;C)fXpg8mbr zJ8rQj_P#3Vqc#-esw=p@T3-$9--S|dUhgDq>q@;~v%dIK22zJp$HV%QtjbQ=oS>dS zUa8$dFGWqKLHdJeD0Z!acT}eHlRlhI;@#2hLVj^|kW20V6QHZ9ZuFc*x2JNcRJZ7~ zRj!q{@_7*coE}~8H=#D(dzMu{nHBPQ*IJQnY!*~IS5@D2+JvA{^;o)`suXo2`%vAl z&MheJv(;6zwlohuFG!-Vh;QmE`T0KdnMo9G4tOZIj6#?F-m0e#?Mhpaw4%6d))-TT zs9lO;m^Hu7yw9{&I&84}y=}FIYG6`S4yj=m+-hx$*Krj|k2;Y*mj@GtreWE5UB4D4 zh&s2XigJk{`QudqfUCW&QFW!Pt;$&#d!T`2c~Ge4JjY9{Zj7hO7#wAstzZ#onU~&TH`8X? zc8&NU3EtEP*n2HrXbRqR2o6)JB8{gjaB7<_4{Kt=j|hHXf*T9jrmSP#)6x?IzLhIx z>&j`Vyxfg|BxxOot{hT#yC2zmqAkZ~!=BzblYjgp|EL^032hNMv_a#g%4LJ)iVFs7 zOifK~jih~xN6-kc&4V6X%)a$RC43UIR*zu)LT}oE3aCU=(%gr8^B(Z>bg7P+pa9SP30ENcP=p!m;u5Cy8(ly9tUXD=PbRL`hMOu= zPddCosZx<#OM1N2#yIP)LO7xZ6DzZ{s0`d8_f4Wj4U+(2`mJOc)zbOnS4&$UIF~vb za8*&71Vk;Y)uOkB&3X$jCtj?siFZ0V1k~$e{)BKB-vWwXVRqC01=H(NfzXj-4Tht* zGmz(=VHmkv13!9B;6cBReg{yYw z{F^9-KtMI<17?N)p53LE*XYFf!n(TH1>iSz>T}P%wF%njwyyN5;BbG zSIe6dg@=dk=lxZ8dLY-)>`WYU3xmb-j_kh!{E%y<)$8M-WL*FWV2yT}c7y|e+*=Z> zI>BPRNBx{+fN*E&^foxIGqG`r^v9CLJwzFO4_zuhr943$q@D)wHWyh@lUd?bu;$FGX5xzQb{Jn_yEe?Npy*{ zn6NsdS!JMH;BP-zmzfOHuW6sjWj{qB)HpW7h@=A& zwwjkh-jM;|olQrPvY^Ne-E7_L2+Mj7;VbSC7YQj|p=YJAEiSitTlofpcC8q}A~r+J zm{u&E7 z$s=7?gg+DTH|+c7`@|C@O8Zv?9C=ZBSd{7D&%LE(4K1uiD&s zC`64W8E{#Es`wTDT)HU_r5Al)BWR zTX#=J7vDh7rsv>N!ZwGpa8%oXTqdZzM7mVyM4%?5x_^-#)yz;=%!&7$@0M-9xEW?q z*8Db|1NjpymqxA{--{ca9^ZP1oHoWs()a?n>hHZo;HZWTpcdJ-?OAonU{$0Wj%wO5 zo}C}5{%3-RMi;JlY)Nk9%Ma~>-0~n1)A0QpFAk3{HWP}EpHo66+h+0$M4)h2O+mpF zjt4ORg*i8*>B)i;Sdj3@H^GShr~A(z{*yCzfq{7j#)9}b{X4rT6bF_S!ns#cj~odv z`xw-$>H(#7+>K1fIQyj|K^!REH>cQ;+iF8Ph-sk4TE%a+5CUXdW|dHe`#ZyWK?j); z=-@L!1!T+-6Tunox6F;4z3y@7tv(!d>EXh8t89R>T8iArGu)%Uu|WCdgt^1>qLi$G z;DvV0h}qm%BJk0b6&zyBi%W!sv?S8x&7pFc^rP(%NMd^FnVy^&mjmad&O*P8ITk_n z!YJRI`F?9e6KVj7yH*Cue`4Bjh}RNDJzfCek(HD~(c%bG_5mq~l*(!=N?vdX5oif^ zl&oMD*1l2Ygu`S(4ODY%VHaCUaZzKW@pt zDG>AO@OV|`6)+08FYG9vaQhVJ@%S`*q7h{ng%zQd34n>Q>EcHW3(@~rceLe>ymQ$2 z6nVV{w_%1UV1H;+t1>M;(&xT>tFTfxB1~~lbRanA2-@aiyQlF64Yax6fieq1s<`o68ud(D?H*PE>teV zgcG-{irJZuj&??OZO&e0bj?V5v8+k?j2XowVab%lRSoUUR41PR1?~KHK~}8z=Z`

      {DT4YxeTdea+_hs;!y1S4zm{2U6{s^QW&jr2qTi z-0NaaY7E4HxCg&o|8|bn6W&KdzL++d!Z-Kaq`3J04~aki(LSD!opENgcik19?2IgJ zkInCO8ggkKolv)6we>z4>5)pr%-@{O?w@gHIwd0VlO!k4+(mTk?I@BjEVawxNJ=GA zI*&L$(}rL_=po28(Voak3s3j9ce$KDS_x*@NRI`V8!FN}H=SXcOgQ9NH zi}C9cUJJbH=Au}^8R6dsZIdo(bjgU<=j>O^<0yD(5MZh8pe_N-yy;Q<9~ zv{aGKrU~U|t~#U9?~j4P5*(dx(jD#qHW?1~dDAWmzQ1_->`_diKNdQS%OhLL4%;o3 z%bD@^4S?9wPCF#vHfdY5cZ-1pCBZ{#u}rvHvAB)%DSC|H?wT>4yZ|sTM@{w6<*7LM zwG%5OKd*E=NG66^R+m1FABqNIM-kiJBSF?+WS7SAYo?{6hETQPzC=^J{{Ar=b)3eF z2{)n(d&6}QzW0Eir$;FEVD8nc14(+tw@GF9{4b?gS16^v>8|Rqh>3oM$<=3`NWHx{ zFvV;{%T-aoaFd7FA>K_lr8?T>*@RK85+mcJ$y)5^o%c6FWqfWCNX9IAONJ~Mu=(%Jrj=AT0SKn2T9q3fcO|SY?-JmeyjNhC)2m&KzWqa4` z8>k{M2-s%pD2(vv7@Jr<)sxCW<%XVdMc$$b#;WY#@w zO`+B<7a;I<6j*d;KORRt1?qwvD4DKaFfr1KMty~VvQEDtL2i~05|pX&#dDiu2E~!? zR@~ks)3np-DgeuV{*&i8_wkfYz80XTJC;S#QGl*H4QgKOp)Rq6w0c-{IxtH(+m&&U zq&aX9==}V97ghyhC8s;4p;u0g&g(E{bc^vV*L41@J`60YPJScRSl!YvE}4JjfT?h`)f4r?a~V&A~)t<05(=DoS?} zezGb+1S<1+bq?A`g2uVKzcK%Sj zEIM+NTQX0gp0U|M#OXO!9{Kn*cd0mfKt28k^W-eb!h;2t5IrJ6cRn1KtG9;X(CkWV z9S$(>P~8KunxqG60O^h)jgB`M?WFF}5wWCwd)6xAOM)7G-NJnFwQ`BM@c;vgqsRy%q;l41XG>yrA&e%>h{_N#_Mk<@QK0Iv1OuY1%f>ZCa3w5htNjBeJcjB)j zXqjjG1%@3>Q5&^dsG;Bc<7}uT_`fQZ$CRK18Y6ZD%cfmP?f0% zAghz98jV!yrZz}dg|i^pFtX~@CQ2QBn)?`V=hu+7bY7fQq9o6<IU3-%E()Y+7Kw zg>CEU%$nqcnb*q*-|wmN<)akT!Q;cSi8PWaptFVbeQSiSx%Qpas_H<-n0`R+V&H_4gEd4`uwW&?{k=rLrrV1+; zkKwOt#FFCh-iH9M8P322H5yE1*^>>MLbEEZ|C2c4>i2_4D1N#V)~(hOa8 z{;Yrd_EdO3U;R}Yao;w*#^&s7Gpe;oETae5TRp(F!u=&;Fi}>UmLAfUY|ukr7nsf+ ztUq@H5sFlMgY2wVjx{;86SX17UKDfb!(1FHh$AI2HTIlYo7Pg3s7h1pbuUjR>r=$c zywx@Z2S+=!rYx*bn?D!KR-fCjZ~@T-XVI$j6>5~o1>uglv=siJ3=6g&=Jy4MGAW}Kfzi5U+TnBuLm%RO;pUmn zuho=XarMDw^}A(V$1krRzmgOm^7QXR7=6o1ROw;ek%cA=p-Sy#U9mzi0YEZ6S>pa% zn5rFLtjEq=3Emv9r__rRXFh*#f7iM^tr()xTD?QLxl>XNj6-0kFhEG&6KUUyH`uD{w+7S{6eGmXBBrudFZT*EL5+wYQ?*2J4w_VeX8_w2otg4h$J}ine9Hz$ zx?u`K_is*wH7_+#kJOdKZxSmKUf1S=s$CYzyrZg1Ozu-b4k-}^dYCcRATC!*h&`ZA z0Tj6{C?lf=ePc(bS#=3tGgG6qAA8~GV8R(q27q$4^Gu=y2&KJT3oOP>3f&v{n4=dM zPndq5p5ePjpz*QOFa$!7xtmTP3^D1A&WddPkd!|ygs|#mq*k7T@%~djIY@c+Sve;B zXtLS97yhIVl}XMSkSB>zC6s_zgPeKXfI55Nn7{j5SjufYv>i0oiXj6Jv+E|S`fnmR ze>$%Sjo6~Cje3eus<{zBFBMkPURh0-P9Q+Dn=u-CxB4tyC&}Biq8A`%WcwLGd&b!ePVD>uoOYFMA}lUH@UjIsgt^=ZV21*d3sJ0Zoz# z6JHTN*GT2e%O;D{&xE+X(_m}Y=w))x0!wQCYy~20tLSN#yo^8a{AS7P%M33o)0di+ z1DN~-s-;Q+(()SNMzrw1uKefvPoOtyQ?SJI3{1?ZDxHqXLdvtFA9 zPaI83bB3uTOf^^?>ifl_2CjU3lV&)PRpmYed$`nn8m4Eyd~RkbVql;-^$%1Y5O+8C zP#x@kn&2VDm<_4d65+U2_$XD2`xh7LIqdGM55QzRs}6q8sIjj|wLOTMOwJ(ZOEyMo zoq9==)a*t|mv~Uxv$g^QD*Yz@0NI>gBQtLZF}c4``)L#O+E(1w zotz@zs8v(pwLyI8gzC=9zb|(LFUlH>0cGiXZ?jgnBLsWZ#C^fFVd{xrK@_gaspI*6{W}r1I&ZVd3gsz`xV|i2L2170m6dt@2RSL zlhKT>MvqK0)s%aWQwLL-LfCVnYek66%3ti_v!aBtxTmq|MyouUz;`^JIRjHHkLc!S zm4cI*+8!U!2b+HW*x(oC0rOLGcz)_T{{{eSv8x zomJ>D55-W*WQ?`_T}LTB8K6Z&3Vk6yx!!62^hcGhe=(6RVC5`*b)tNYb@4&SZ zfai@>Buz{7QL>g;$>j``S#(!5;Ge1sLQ{0e3(k&qIAY4qNC(O6s!JHwakRAWmv#)- z@Lan0kC}GSce;9mp8oy3CGpfFvwT7tHIV`+R&J}1>F*AKShq5yIhrK9h&oRh{X*2r zb(`#s|MX%1lQ)O5ks$*hxZ6rCV;}!-D;V`T50`nS2dL=zR|SV)PVihtw+w;*FO>6m z_c|lGG`?y(yi8^NU+k3+H!>f4l zYi;f@m}V_croyMJNq%u6vWi+TCLEUxu#y;=2QFGX8E#wxT$Jou4C}Sk@Ouh0}T z8x-)s zEoQJp1n#ipmZ?aF1M=X_!_FGDh}Wz}WWa)CFL%8h+0hfwLhJ?mi)IX{g99B>#IQ{V zRL^Nw&1HZTBY@_O6$?#Yh}O3eYK#a%fs7j?1*LvfKpl(OmDlS3Ru%}9KPumV8EYYIouE1 zV6FfG3DmQeU9T9s&mrj3P*xgFo#!Iupl|3lTT{s(TPPzeV&IG zh4_I(9Rko;e1{Lp1SCS7nhxKaq3r#n(ZexeFH23=asyu5ry}cQ7Z427@>Dgv4yc{6 zrkI}>^9x!{{-T}@IoWdy?_ic9b$c6&AfhX7;NP;b)yCM@;GJr++C89oupjkNz$GaA zAb4Hs0Rm{-QCuXSO>`UY65kwWDIZYMl#Uh|N>ns(Cspi^yT##N5fgo{^&7IAc+@95 zn#1R{#K24RRw$6;*+yVFOd>$^2KhG(FRUYj3HDGXf)`CfxVz%vuii}iwK)nIlNwr# zb8c@CM;qx7T}ybJn>~>}*v;5yD_V1pgY5yhl`Y0Y!xPCy!5<};T40t#VX~~1^&@uX zxoyr%j4-g*p7xt>$-oMN1~7-vUqG?mqs?|eM~_+3O)9XTsIJgPNtV>aYmUf^5%Sve zsIpUd9<+eOf!S=joQ3;12Y1!zNmD6k(@$ki=wQ82(xpcc@%aHc&y z4(*&J0EwDY4cnnd?&4kmpHqx<#jUZ?_HMa|Ae zkQ2`pz@0XjVJWG(^VP5;7kHLS;6Laf;<@r<9WwX{TxI!l{z`}Rzrd(+a4)r?REgh& z@ti2gox%Y4fQev`zR{Nelelq`qdV)^#QR~mClcr3Ud6I2Riv_65k#_zsQO~=UA{Jy=p-Yxb**y=&ckPz^{|B<0x)7zB_}}uc)oNeqvmY zV^q2Ky29=K`Tm$C-L}q2Lv7ckV-r?vGT{Q59u>I|`C~!m_VryIdtgv_AKMzH zNIUfgR`)YhybW>q+)*LEV0^ywhm3~wK#9XpAbN@Y4FMG7s)cR`S)GDK6yJxavyY|z zX@Os*`CiV~3IPP0%r`87dMSa=?FwwJ3z!vKZSbzHz3GP-=;6r5MhoSLAGcQ>-HZeuLCsd*N#P#}r!*tw6Js(yd z;nZBOgl$`)>8ohU;3#9GjM9w~n^Z`4;Z#2fuHuZM$DOAK<(G@#{l*ZGU8w ziMR*uTK|oT0C{xpa=nY&Aiun%Z#flpI*?)98R590c_n_4y}~B{m*wKZYlg~Mcha1e zNLG45kU{lPq3(9x54;P|>X(XDPGi)X1aiyBuG{JJBbJWnN>&YIS2kdHFYti4LY^v4 zJxjFHqAwdX4O^hRov}uK4y+_gA-}AFZ!qF?I-E!%GJMCyt6^LK)Qw&Sbi!GEP(mk zE0Q>VXFG%Jj*SbFx?Z(vt!OH}jAec;Uzn%gL0W)Ycd;Op9(*aO6b__@NP$K{Dv_#_ z(pQnX%u;?npusjtk)68>mbMkKUf*YFuW;{a4O2#MeI&=4l_YJ)%{HalTYj0hzIu9e z6v;~ZgD1nHzx7Zr9$e0HMWVUk-vRfB6%$`hcXB+3DMZ!yB6;XUSiS)cVJTXiiKr$S~ZmK`z5M`+pEBPm2`vN=Tg-vR(6_Cjr=wy^@eZI_IWpF;DKmdb{ zk~@gSL+E^&&>J@S7w=yaiu=5f2`q`*=yiWBvbE-@u0&~41@eIN2x+vQTF z7~|g)7D>NphGmKy$Rxap?|pjk?{lBsC8bX)7~^c(9Ie=auzS{VLP?Y`sZ7Pf=Q=l1 z7*ZVZ<0G#W=_?2`GJ$VGolk6-VnC-oLa4LX@jQ-EW?yIn&A_5&K|C3{D$4CVUA1`d zf2u7lXS22@{8tW|{L@?iuf@~L*=aXG4E0l3xU<&4cc}(}-sGUS63mMsHnpdlCD!#D zG5&~&5!-HP;J!dz!6z=0=OLV20gN`t>I0yaH@Q!4@nujj*tfZTQ_c}NY~qXn*_G9{ z<6DW6s5iq(9hns{J<8?n-?NgZb-=)ghPvE=zA-Gq4efIQ*Ul9?bBZM@O++sM#N9M( z-|TOPsc2P~8#+S`-=kzB_L0R&7B_=#p`6EaH>HWFKDr;!|BMRG0w0oTU;u#s421ta zDsKLNqr&8W8Wkm~GIB?v2tAKfn%Vq=U9Vf>XfN^G|F5&N4619{);0up5AMO8;4Z-- zSc0<%9^Bm_NPyt(?iSqLA-KCkaCiAu?x}qbVV_f9?w+$|!CF6_>gt;ByGD;5-3<+a z{w4#@c^^l%$G&QkaGg=!U5fr5^1aI$;Xomq&pe}P1_CPcjONEuKvTgke`v^bPPR^F zoiW06&vL$2D7t$H|Birxl=+BRz;)QtqrG3x_vM*%af50KckPT!!t6nqM8pM(DUHC%TtRPG=vAezFlQW=MJsoRB!sBESe~r4=UXNmcrH@DujR3H(imR z*`_Rw43I`K8ulb0=bK3Ym+ti;aXN1%5AnP8?{{{WOm}uh(xvzpyOp_}kPDEd-X$e{ z+QX@QKWiX&1i2p0UU>G^pL*@5t}{lR0!E5-xNF$kNC#`yV${SU$aX8^o8;S1y)@Q~ z*D5t>%#^2wl1K4z#vtw5&Ehf zu(p%9);J@dsio3G5+8NbX;b>^uu0ib7!(5+$oe=8?3BKdeLS z78TKwijxnyv?;AFrO%SyTQr69$#FYcMY>KDf5G&&!+Rosx0e#QI$=m8ADx$DCf6d` z&x8d9@wpTcB6?R##Jt_+OPF@48A^G^f|qw8FG=Of1eVLmph5nXa+)7fbl&#|oH+rg zGc}uJOWBk(l8OG2KmZqMN%QWmH_a`(SW^krr4w7Sw*V8p-m83W<~E~7VI8=ftu5rC z#FKE5bDYI!6^#DmlBnoo33Lar_%+mVRQ;CdZjDg^)_MaH{LOl+dQ)etJ`fS}pcv$& zXne=7Q9V$t&`%Jb*|c!RF-?Mq50-59Px%m-x|i$wyyo zKV>gM{}bYd8FcZQ5&YixA*rLn@zAC3p&{8c^awESWNO z>T&RDQG07CLFqNY%^B4j19n%`Y(J5C^4?@--O6bh7r9} zfstXDxIrpI@hbWFr|A_RTtE|<4WtOf~Ntvk_LcN)e<=j+mKVJg=-G4v{C9<**6=WII^&CakUFrqx6-#Qc`SsCxdH zkOC#{anL!)P0(D>yL+#JdZ9r(>%UiFfCkLatKBa@Cf^=F@@2k2&Rc9)(Z1F&qs%u zde~wj6!DfN&{M}nu3kD%2?`7Pid=LQkZFGUaVI||qD8RR5Cb(N1-CR*vL~p5s6cH@ z^a9UR+Qg!vHMG27#>8<(mu7EhiV*FjR{^*Dvm7n6O3epYwJ+No+!xA9J+c`FKM|Cz z?C-c^5Rxm!4{hVn7+L2D_y!-FwiN+7B^kD|LY%X2`LbF+zP8@dvH>|GYnVnrcz|#y z?TR0aM~vO=Q)i20r-5SIRjySt@qde_BD4Lbfvfu6XYSIqUX=;fFD(Q&F1@gs2f5W4 z*nT}yq=@@s2Zj$Jd|mTglENO9ZXX?M5U4zV1d@s@llSfi5=+J;S~RnVZ42#r_p-ib zuwc}xe4zH7P&Yc;MqV4##P2ps;9Sd!aoKx&_FXMx+5#)k?!KM=aST{`5UT0?!9s1% z1$VSuc!ShD2!|bA6j-G*OLwOt;_~LA)2GJCMANfQ?Bcl5Wv)qRn`6HK_VTTS(E?b? z<8a~Ldvl>fzwF*O3I2}hIO|V>6?wN789edn-BPx3`c2d3;&2=J4Ca25%!%c1$LGtG zh%i6x4t0QP7>$e=hx$vlmQFTzG_jpzr5!6E8GM_XOtE!T-DJ68P2T+85wkY#S#ZC* z=y+f+Fll-UMgbQ7sFX_4FlNGU=>O+v8b9ETf`rG(lsuZk}CZXiM7F}OBJJc!7$?J@aSJ^-hFT+vIGVezcl-w)l32Z zQq4Ahq~kVPnqh=C`hnX7sa=}m`*pm-FgllF8XqU zBe)IOjvx6RoE0595f&Y~5=76Tq#jSfy3mOzZ6ysEy}&zX79%s8c{&1!H$f;?-kQ$n z7&4~6k8$OJ@JJ=bf^__+$|e)g=;yK(HWOiLf*FmXjt3vsy=f z)*Q-^u(18%YnmdY!%tcb_iM{huJH6cyr)Y7rLyw!Wqrm zSd(aK81&l58SS~Au)y$0h-Npv3=XS8ib#*_)1A#knM*|I95YC_uk!#4hQECoOmvR+ z7fWKfN%{Mv|q^cW-D?#UhF_$mcy0kc-4W7R7go*?wf$Q_EAg&Za-mwnYbPB=PCL!<1n!AwDd#dipW26KY5Rt5)_rV}lk!0w zNCac}aUn_5BX6q&XK-me$TvF5&VEp5nnOdVMAf$9tHU%trWl(5;!cSEVH0XdhN@;O zd%1`Mvcj*TnDsd6U6+Y&8DM_&Jx+>hQSDnJ7I?Y|MwLSd z4K<0B%URdsL!rgCvhoK0Mrw0??yj$DsPxC`GP58OJ7IA+!DG6RR5H}jN`=agG#vht z9yo2ZL-{!M77Gmv-m&go(wPS~R99fz*psTuw|YJ1GOH;rm(!flHY47zrXaQt(ppR6 zo`M+Ba(5crobQuDz_!sZU)Nqpgp`+t?&kV+=e@m>fXl>d3ATI3@x_fiVxS4^jHzb= zT=+F(OaTA-g;++q35{Mp(~fHCO1(q&ln-hU?uCJF$LDdMSDZe)(P&m$;!4vw8e@J4 z4j$2<10>argStk}>rtoSnY;ph;uj&-D^}#8eQS)8B&U{&(=`r9hX9E}FcIBy;dsA@ z0j$2we3$@tQCrMPd2`LRbG_(@GEt1Jx0G5=Lzxgw4^@l z!{L?0%Qu;RFOJ1UYninpC<<=knyT_y9YH2X8J90cJxl^6PBBt7SVdw$dE1aZ1A6l+ zP_wDzzK2!6KP$q7%m34`OPGebfEnck>5oI%c-eM6lF@(T$D7J0m0>xp|GJT!ei6ql~k0I2gs0cvq{%ajO} zc<%HjhUg&ApewoaFwjc4A_VfIJb7$ZI z=YKsf4mfZ6l>%W$ZoZ;|*4hUFTr4H-vNU0dgoATd?r2vgX)>)w(0b^pi!5g=UyQ`g zVD7=^^Oq>U4-*Dg0fcjR92s(KXf4CsYS>BYRzQ+gF+WTqtzGd*qw(to^y3gS1-!2^ z@#WLl?Bh3=`m8eVPu`>Mzr|xeYX$C{E?BgGt@TclH0zbw-8-adHnWkQjk9aMyEmMs zW#L%g_0d;(?qs`lONGuGLDD1F504YV1hj5(c(NBhL5B8BL$NT??%K6_y2BQzeTwg~ zJQNxHVhAvxgd%@67biDd{AJ+oZg%IhVy`KpT2RI)x<$YgYeiZU;nTO^b)2nZvH2^# zNkIHrFClTIgyNe%;gB5@SKB$~nRyR#if$4P=SbpPE6j)xWTXAEU7-cy*=B7%#0LSU zsLU!KAI^@z4?L27H5qin9tzPL@GP~woE(RrheliM7bNgzhQIPavK9D?^F)5QOxk47 z_v4JNqaHGHG{(zfEQLoI`w|Cxd3Qesh=Fs$DlTU0Y|_#8f>@_Q=}xAH#agRGmU1R{ zi{iafw{&Co2yu>Z#zq1G#(gOFd6KUh(<09gTcgMD)RcLy@Ay^5!QD@HS`%B zn53{>Gc+2~*aUp+r~I_oF8cF`g-j(q$<4k!D>3t`u0<;40u=i3k0b;?eTzPC#BS5Z zaqH7>*^cO1Z(}d7L4M;p#diRxN;)lo`^C;gmVyHUkeuMGac5DJTQM-AmD1C2i@LFp z08ojIYbhO%#hwLmVaFM$!`hK|1Ts$76F6KNwj=O$k4xC5tSdvWiXTo)gq)M#IxbL_>+vZc zoT--ES3+*|pzDyqBSlq&?CnDs)l4DEG*P){5@%F8I4j6qO>IZvInW$V^>bsS zf^R_NcV^({wGnV((ZKCayys$hQ^{|SDZ3O|3Ui12JX0|zw7V% z$&DK!_r&s2PI+AhOlQFH6| zb>b5Z(NTmA0O2M?X<710%p#cx+i34K3dtLOOajR*D+e^dqy$@9^cS2SD1ru0Av+h^ zwVbLPMR!UOn_Y2#h@Cr&+dINvW0)j9dM9k4?ff#A|7<&zFViW1c$8kUl8nVVa167p zjLfO;McH%}v#5iM9%MdBtCj{F!?0u+^<@blZW;tVUNxud1;r$j%m^bpaS&Q8SsGjq ze^aO_KFFUlo<5(7pFf&RMHaeGArZ!6uOx%5L4LGj*tOb2`dEDeRwv~aUeLX}KCdEd{Fba7Q6MyxPxI}hn=Ca2FKz^lH0fP>J@_^J_c4>G{coV9 zob%sXZyLUZJ9oi6X>@Zb?DTrgOG`+EhspMewJ z{OCWD1*e7Z1bfFkF^}J#&%4GF(}@_*9$$k%19sIdibz6h?ciJq$wZTkvlnaY{+iR& zbHW8VPF7Y)xKyh^FjocnmJ?b zc%&aMxU@EJAk=xc7CWjluJl|d0@nk~e2D8wK#w^^QAg^?U~`;Q>3p< zks*z6h&3lTcj|F%!PeM`XAe!up}V^Y>9IAAFZ|y8m_~VxmQx`-d22bcD)WZq<_w1H z61UYY=zB?x^!8=Qsv7 z;79RB0JDNKuh~PKi_zRqR4X~P0jFPy^BcAq2wNHz%V9x5cNI++2U-K0WivX&T zQOu4J6~f@p`i@6)&WKrIzWt|>B@rF$XNI4K9QdOV6ol^3DSGKz*zi)XrzX8@@3(l4 z`^ae0U|puu25_e=iicZL#7fc#)5kynEbh0D3Xm(6Ggh!C*dg^?3 z^uyzqfk^O-39SQ%O}}j7KN|=V+>2q;AD$2K%Rqo}OVgb)G8d<(GBDiYoPT3VRE88EWSjRTZU|Z3$2KCj9JrDG}*(7_Q5sBKx3B z)$47t-NC-$h07x+LlZ>oD>Vg3jO*dP5~{(=0S}Fn9#xf5vq+}T0yUwjP1(r2j@fPI z0wBToN8HA6GYGeFJRe%U8-%~GvO$vWe(v*yImM}GB^8o$vb*Pc3=-sY{({UBTD<8) zHAo>~IkE_$PE3bq^Bpww>XVYf5!h0<;d?^?=@ERN^|c>~6d5Gguz^p6KWN&KdVR%( zZTgRuk66#E{hmk0-wKzIjBpJbGXcJ}fM&K*fV5U44ML;+3Bq zDECAo0{q}4t3EcZdZXTe0{_9r@<`#QAKD|fpV_9^wFP+bZxuej!0S&g6k&f{vq+UNba)vC%3LLb8BF-8MX~r|)9~_d4gq|E;U`Yc?mdl)2=PT8)()M3 zcMPL32EK24%Q^GIu65>P5*11X6j0vhws+2P9iX(M9CeYd*?d*D0G}u2-zuXD^{v3n zA!xx?C(tW038QEabQNWLZS2D8gl25v5L#pux;p<2j#T012mB0(r*Ql|`q_wvI{!U_qa9q;AAG*x1H<6RDu zTn=@cZ(6M`5q%OYLmTV0;C3gQLX(D1kV7|dIX~cI52S14l_d4g$S<#Mzgz0{mA-VG z#DsFT+N0#Lnw6VDz8Hg!cM6)Eq3vN-(zN^88$Ng(8LSaWe4mY28b9Eh`Tc{~#Rzi3TC*Z{H=S?Xri?QHcmRc7=rB6tqw?*|+yy z#P?4G3NLaa768B@=iQNU?jPD8^~96nj>8Bx_)x~zE7js+a9EEf%p4D<4kjStPP^kHE{~|zJyVKNV^6dv#52V?j}>IaqSwXgC{mHA*++~^2(dCT zL=n)OkFW(T{flZu)r!vY-Fu(*R4I#u0ixI-R>ksyv+KwHw_ce7hFu|?Jw30jSZy;c zGk3|kPRB{k%JoI)P27xe*1J#P+snCwsyS%Rv+Nq%*~C6$&M!b&k~4PE8JRR_W4>GT zK!!@w4_UiWy7p7XIOtwnxn^4dv#P^3N^rt{BYg3(FlwQG^8wkHzM;?FD623bOHF2N z%OLrLAUSM(a_Ox;taxLPMlAZOG-mVL2zB}1AVd%&Vd=M;V>%^E+aj9=qjsE<=jN3; zY#BXBvWLV-GI9~Ae3R%HGGAavT1a}0vxa^AT(H<=8@Gga^nW%iw!&P+D?2?d`@{s6 zyF_18tG(B5ICJ~9*{60_nK6gnBRFHei22E9?8rNfV8!_{_CQIGUCmtJUC)=o%PH~^ zb~<-_KL(8kx$hxjojm&BdO|3m%LgRg4tA?dNuFBHZnYih-6c;7WQi-G7b;VK}SW(H!{2++$@#&~@ z2`E#>Hubtr6dlDHQ%%s}>0%r%7;Dr<2)#Zk>I9HhRjnv=^zk}P2)(o0R0fLyvpf?YR}uMIx|aXPr?cCIb|oQzxpJTz{}-QrIcogF zU0RBjq(Xo}-{LS8bg5}8@2ufAM?97n8ZNf(Ynbx1gzVRorV&Izyh`4*Hoi#$HV|5A z3^8KW8#!54i~{8maz)A)yUQLLFej$P2&}k9(WBKamsQsoLID<)USU*vRPS1v2K%kn z4->sDQ}#35ax>AxWK$b!1m0z5Td~T8Oi=0=t8(LFZWPA~2HuZtW5Rk@F_Fj}c}JvA zlb&4Yh^ZOb%PT_yF#75sF=DpWK4_>CyS&$vPM5x1;AwKcyJX553_x9lhsR(nKl?#e zryjr?R)Ns4Q?bXTC+}zP1C80r&iz_A2Qp%i8v zlyXg7C#GE*i85KP9&O%;zbIEW{)uUqYaT3LX9PirY=Zz9TijGn0D%jiljywul~RbD zwxz6OgYS*3YRY()?<{#0NMm?I_W*h!<5A9o7&7NO=e@q((j{S$*R&_tTyL$a&2gA* z$i$4I(9c{DpSTOd;_wOxAauO@?^Q^=*?OpBVyykyM->~eJP@=I)dUu7EH%SAJSRfc zlG$WW38-_<8yw`%ygC{pB%_sRn43n;kcQozk--!`1!3P`>$&i41YT!LE9>%R0&{Tw~l+Yz-Q&&&-sIDXH5i1V= zC|iReqjTTFvZ*iDr-jv@7QMgCPvXFAHx8x|aXj<>J#pu_tu}PTj^K2a+dH8w{9N8! z9eM%uL^I}B81Jq~b9vq)nUrNAi{RxKl}1?3ygTy}NZNVv_a%NP zXu8$Q5+*l44EXg_)Ma1mYpZ`CYst7}Yq`Al6oik|8H-yG*ei}-UhW(T9WL4!sjeoS zfu^1=5F`i-Q7zfL2srFruTJz38)XuV9B|szzt8U3U^*q;Z!@ zlJF7wIq?CK_GJnSy|^%YH*ao4Uf|5TAcovQ7l3oW_M=LaF#*=O#Qju2tltQ zG?ljW>_GW03cqlbcM2^2Q|0zmrUY%DBri%L@sEQVgam@WOhr|F4tL;Hi1QGUw8BgE z(kx5c@9-GL?h&QcrP_%lazU!XLG=l>(u1q<&`#2$SH0IxfOrT#R|_i{?SzM7hr`dU zS61qK`Zdg4YGOq*Q?u4j{TlzN-Z=bEcrNM)Y9fX4GTxos z*A_FBQ^Td3->hQEgz4e>`mEI`F0!JRWmWjz=W60xF%H%&@LYfDKEmtflYzET8=t4) zx&u3lYeNoxP@#a*EnSkDkkedFk=2bmfRO%8amfTpZcbcP{%jPs~o~TO~SmUTN_Ey$PBSjaCpldC3 zk*d9m0*C0SKK{T1Dua#yJkp8=SK!u|DwuwpMLKob&aks<1>c%wxAD&ye`aKh)WU>8 zc(+8w_K>XHtX&GK20ETNn_7X3Q~T%>$FCb5M;D95e{F*ZN5WawSE|KYixlc76`zjv z<=Kg?h0PjgEFD9H61jlXXFBHyGMMz{>-Oem>o>s;t*0$3tZ+Bmr(@c4aWo#09;w$G zR7v7zZKMRyQChKwBx6ztMm&60f25p}G25qUlAEtQCQN(8&LC6vj;@+uhp(Z(yA=0dRs%P1a)|&;YX346IVsRr=wRp| z(7?O2*jn}E?NJ>ZC=}+s6U*a(@QIQUSppd`e zje+0%8~=xeKritvNzw)wKv37;@YSz=!~a=j!b|)rfm16p5RC?e{oXm$!0!Ba>;G`( z@g;t@$*CS1h_Cru>l=W<UZFCcp1 zZ=I6@e9!*f`ajG_yu@b@4okuU@#%kSy*uz1^WXSC45VM;Eg<*hNr7m)zu~um$(DcP zU;gI)mr3Q9_#&!T(x*+iGQ?(MA-a2*z$1ti4FVViF|BL>!>fx6i5~JIs{t5{M#F6SHI*0Wy z^gry7ga-6a0>D;Z{jJqtBL9p2zZ<|V&>cgPdm77u=nkfrtxi$;3;kUDKY?Wje?2dv zvX>q5{1oZm-;l!3f1&@O@SmTX^g9~U`7iW86#etfNPkCr2K|Nphk}2;;m7Z2zqG&5 z|4{7D*VFutcKiMp`X37YdHwybMC@NzmX}jI$p3kH&dZnNc}eZ>^sMgZ^na$w=Ouo> z(-{Vy(f^?=pI7kwPJbABPXE6v^H<#Re4_q$y6gBe`g3vqW!=C2{(r>zd=~F_`q}g| z`ajg?^R)i&be4r@^nWPP=gFww>0T?(=>Jfq&xftQ)2X(e(f^@PpGT^{(=~UW(O+uy ZzYK)sq#z)F>5%| / Recording site / Year / Month / filename" - }, - { - "id": null, - "textContent": "For example:" - }, - { - "id": null, - "textContent": "D:\\Vögel\\56075, Karthäuserhofgelände\\2024\\Juli\\20240704_0130.flac" - }, - { - "id": null, - "textContent": "Library Format" - }, - { - "id": null, - "textContent": "The library can store your audio in a lossy or lossless format." - }, - { - "id": null, - "textContent": "Lossy:" - }, - { - "id": null, - "textContent": "Audio is converted to Ogg Vorbis. This is the recommended option. The audio will be virtually indistinguishable from the original in the application - both to the model and to the human ear. The file size will be dramatically reduced - to the extent that a full 24 hours of recordings will require about 1 Gigabyte of storage." - }, - { - "id": null, - "textContent": "Lossless:" - }, - { - "id": null, - "textContent": "If the lossless option is selected, the audio fidelity (bitrate, channels and sample rate) will be identical to the original file - even WAV files. By using the FLAC format, the archived files will have roughly half the size of their WAV equivalents." - }, - { - "id": null, - "textContent": "Spectrogram Preferences" - }, - { - "id": null, - "textContent": "Colourmap" - }, - { - "id": null, - "textContent": "Choose a colour theme for the spectrogram display, or create your own. If you select 'custom', you will have the option to set the colours for peak, mid and quiet sounds according to personal preference. You can also adjust the mid-point position: with a value of 0 or 1, the Spectrogram will be two-tone. Values in between will blend the three colours. If you set the Mid colour the same as one of the others, you will be able to adjust the contrast in the spectrogram using Mid Position adjustments." - }, - { - "id": null, - "textContent": "In combination with audio filter adjustments, a custom colourmap allows you to enhance the contrast / visibility of calls using the colours of your choice." - }, - { - "id": null, - "textContent": "Window Function" - }, - { - "id": null, - "textContent": "A variety of windowing functions are available for the spectrogram display. Each has slightly different characteristics, so changing the window function may also enhance the appearance of the calls in the display." - }, - { - "id": null, - "textContent": "Timeline" - }, - { - "id": null, - "textContent": "The spectrogram timeline has two modes:" - }, - { - "id": null, - "textContent": "Timecode" - }, - { - "id": null, - "textContent": "displays the time elapsed from the beginning of the file." - }, - { - "id": null, - "textContent": "Time of Day" - }, - { - "id": null, - "textContent": "relies on file timestamps to determine the actual time of detected calls. If a file's timestamp does not reflect the recording time, the reading will be inaccurate. To edit the start time of the current file, set the desired date and time using the filename context menu." - }, - { - "id": null, - "textContent": "Advanced" - }, - { - "id": null, - "textContent": "Send Filtered Audio for Analysis" - }, - { - "id": null, - "textContent": "By default, the filter settings in the Audio Preferences are just used to modify the audio in the application interface. If you enable this setting, the modified audio will be used for analysis. This is usually a bad idea, as it impairs the accuracy of the models in most situations. However, if your recordings have a considerable amount of low frequency noise, it" - }, - { - "id": null, - "textContent": "may" - }, - { - "id": null, - "textContent": "be that modifying the audio using these filters will improve the models' ability to make accurate detections." - }, - { - "id": null, - "textContent": "Context Mode" - }, - { - "id": null, - "textContent": "This mode is only available for the Nocmig model. When enabled, the model will use the surrounding context when making predictions. This helps mitigate against false positive detections." - }, - { - "id": null, - "textContent": "SNR filter" - }, - { - "id": null, - "textContent": "This setting is also only available for the Nocmig model. The SNR (Signal to Noise Ratio) filter can only be enabled when using the CPU backend. When enabled, Chirpity will disregard audio segments with no distinct sound event. A stronger signal is required for a prediction to be attempted when higher signal-to-noise values are set. The purpose is to deliver significant speed gains, however, the setting may cause Chirpity to miss quieter, more distant calls." - }, - { - "id": null, - "textContent": "Debug mode" - }, - { - "id": null, - "textContent": "If you have encountered an issue while using Chirpity, you can enable debug mode. When you change this setting, you will need to quit and relaunch the application for the change to take effect. When launched in debug mode, two windows will open, and each will print debugging information to a console on the right of the window. This is" - }, - { - "id": null, - "textContent": "extremely" - }, - { - "id": null, - "textContent": "useful if you want to report a bug - as you can share a screenshot of any errors that appear (these will show up in red in the console)." - } - ] \ No newline at end of file diff --git a/index.html b/index.html index 9b9fc6d..d68ab7c 100644 --- a/index.html +++ b/index.html @@ -1010,6 +1010,16 @@

      Audio Library: +
      +
      + + + ? +
      +
      diff --git a/js/i18n.js b/js/i18n.js index 12f0aa3..fd38fce 100644 --- a/js/i18n.js +++ b/js/i18n.js @@ -932,7 +932,7 @@ const i18nContext = { pause: 'Pausa', analyse: 'Analizar', create: 'Crear', edit: 'Editar', record: 'el registro', - export: 'Exportar este fragmento del audio', + export: 'Exportar audio(s)', compare: 'Comparar con reclamos de referencia', delete: 'Eliminar este registro', location: 'Modificar la ubicación de grabación del archivo', diff --git a/js/state.js b/js/state.js index e60463a..00c1425 100644 --- a/js/state.js +++ b/js/state.js @@ -38,7 +38,8 @@ export class State { this.powerSaveBlocker = false, this.archive = {location: undefined, format: 'ogg', auto: false, trim: false}, this.useGUANO = true; - this.debug = false + this.debug = false, + this.fileStartMtime = false } diff --git a/js/ui.js b/js/ui.js index b628207..5c7c5f3 100644 --- a/js/ui.js +++ b/js/ui.js @@ -1719,7 +1719,8 @@ const defaultConfig = { limit: 500, debug: false, VERSION: VERSION, - powerSaveBlocker: false + powerSaveBlocker: false, + fileStartMtime: false }; let appPath, tempPath, systemLocale, isMac; window.onload = async () => { @@ -1794,7 +1795,8 @@ window.onload = async () => { // remember audio notification setting DOM.audioNotification.checked = config.audio.notification; - + // Zoom H1E filestart handling: + document.getElementById('file-timestamp').checked = config.fileStartMtime; // List appearance in settings DOM.speciesThreshold.value = config.speciesThreshold; document.getElementById('species-week').checked = config.useWeek; @@ -1896,7 +1898,8 @@ window.onload = async () => { useWeek: config.useWeek, local: config.local, UUID: config.UUID, - debug: config.debug + debug: config.debug, + fileStartMtime: config.fileStartMtime }); const {model, list} = config; t0_warmup = Date.now(); @@ -3706,10 +3709,16 @@ function formatDuration(seconds){ async function localiseModal(filename, locale) { try { // Fetch the HTML file - if (filename.includes('usage')){ - filename = filename.replace('usage.html', `usage.${locale}.html`); + if (filename.includes('usage') || filename.includes('settings')){ + filename = filename.replace('.html', `.${locale}.html`); + const htmlResponse = await fetch(filename); + if (!htmlResponse.ok) throw new Error(`Failed to load HTML file: ${filename}`); + return await htmlResponse.text(); + } else if (filename.includes('ebird')){ + locale = (locale === 'es') ? locale : 'en'; // I only have spanish so far + filename = filename.replace('.html', `.${locale}.html`); const htmlResponse = await fetch(filename); - if (!htmlResponse.ok) throw new Error(`Failed to load HTML file: ${filename}.html`); + if (!htmlResponse.ok) throw new Error(`Failed to load HTML file: ${filename}`); return await htmlResponse.text(); } const htmlResponse = await fetch(filename); @@ -4801,7 +4810,7 @@ function playRegion(){ break } case 'auto-archive': { config.archive.auto = element.checked; worker.postMessage({action: 'update-state', archive: config.archive}); - break } // no break so worker state gets updated + break } case 'library-trim': { config.archive.trim = element.checked; worker.postMessage({action: 'update-state', archive: config.archive}); break } @@ -4815,6 +4824,10 @@ function playRegion(){ case 'lowShelfFrequency': { handleLowShelfchange(e); break } case 'HighPassFrequency' : { handleHPchange(e); break } case 'snrValue' : { handleSNRchange(e); break } + case 'file-timestamp': { + config.fileStartMtime = element.checked; + worker.postMessage({ action: 'update-state', fileStartMtime: config.fileStartMtime }) + break } case 'audio-notification': { config.audio.notification = element.checked; break } case 'power-save-block': { config.powerSaveBlocker = element.checked;