From 94fd36cfee29a67be19308688f5a537b1db02705 Mon Sep 17 00:00:00 2001
From: Alexander Metzger <alexander.le@outlook.dk>
Date: Sat, 10 Feb 2024 16:49:38 -0800
Subject: [PATCH] correct setting split

---
 .../text_to_speech_settings_widgets.py        | 43 +++++++++++--------
 recipes/VideoBots.py                          | 38 +++++++++-------
 2 files changed, 49 insertions(+), 32 deletions(-)

diff --git a/daras_ai_v2/text_to_speech_settings_widgets.py b/daras_ai_v2/text_to_speech_settings_widgets.py
index 499b502d2..70c217b93 100644
--- a/daras_ai_v2/text_to_speech_settings_widgets.py
+++ b/daras_ai_v2/text_to_speech_settings_widgets.py
@@ -141,24 +141,33 @@ class TextToSpeechProviders(Enum):
 }
 
 
-def text_to_speech_settings(page):
-    st.write(
-        """
-        ##### 🗣️ Voice Settings
-        """
-    )
-
-    col1, col2 = st.columns(2)
-    with col1:
-        tts_provider = enum_selector(
-            TextToSpeechProviders,
-            "###### Speech Provider",
-            key="tts_provider",
+def text_to_speech_settings(
+    page, include_title=True, include_selector=True, include_settings=True
+):
+    if include_title:
+        st.write(
+            """
+            ##### 🗣️ Voice Settings
+            """
         )
 
+    col1, col2 = st.columns(2)
+    if include_selector:
+        with col1:
+            tts_provider = enum_selector(
+                TextToSpeechProviders,
+                "###### Speech Provider",
+                key="tts_provider",
+            )
+    else:
+        tts_provider = st.session_state.get("tts_provider")
+
+    if not include_settings:
+        return
+    col = col2 if include_selector else st.div()
     match tts_provider:
         case TextToSpeechProviders.BARK.name:
-            with col2:
+            with col:
                 st.selectbox(
                     label="""
                     ###### Bark History Prompt
@@ -169,7 +178,7 @@ def text_to_speech_settings(page):
                 )
 
         case TextToSpeechProviders.GOOGLE_TTS.name:
-            with col2:
+            with col:
                 voices = google_tts_voices()
                 st.selectbox(
                     label="""
@@ -208,7 +217,7 @@ def text_to_speech_settings(page):
                 )
 
         case TextToSpeechProviders.UBERDUCK.name:
-            with col2:
+            with col:
                 st.selectbox(
                     label="""
                     ###### Voice name (Uberduck)
@@ -232,7 +241,7 @@ def text_to_speech_settings(page):
                 )
 
         case TextToSpeechProviders.ELEVEN_LABS.name:
-            with col2:
+            with col:
                 if not st.session_state.get("elevenlabs_api_key"):
                     st.session_state["elevenlabs_api_key"] = page.request.session.get(
                         SESSION_ELEVENLABS_API_KEY
diff --git a/recipes/VideoBots.py b/recipes/VideoBots.py
index e4e74614d..727566611 100644
--- a/recipes/VideoBots.py
+++ b/recipes/VideoBots.py
@@ -346,6 +346,10 @@ def render_form_v2(self):
             st.session_state["tts_provider"] = None
             enable_video = False
         else:
+            text_to_speech_settings(
+                page=self, include_title=False, include_settings=False
+            )
+            st.write("---")
             if not "__enable_video" in st.session_state:
                 st.session_state["__enable_video"] = bool(
                     st.session_state.get("input_face")
@@ -355,12 +359,27 @@ def render_form_v2(self):
             )
         if not enable_video:
             st.session_state["input_face"] = None
+        else:
+            st.file_uploader(
+                """
+                ###### 👩‍🦰 Input Face
+                Upload a video/image that contains faces to use
+                *Recommended - mp4 / mov / png / jpg / gif*
+                """,
+                key="input_face",
+            )
+            st.write("---")
 
-        st.checkbox(
+        if st.checkbox(
             "##### 🔠 Translation",
             value=bool(st.session_state.get("user_language")),
             key="__enable_translation",
-        )
+        ):
+            google_translate_language_selector(
+                f"{field_desc(self.RequestModel, 'user_language')}",
+                key="user_language",
+            )
+            st.write("---")
 
         if st.checkbox(
             "##### 🩻 Photo & Document Intelligence",
@@ -383,24 +402,13 @@ def render_usage_guide(self):
 
     def render_settings(self):
         if st.session_state.get("__enable_audio"):
-            text_to_speech_settings(page=self)
+            text_to_speech_settings(page=self, include_selector=False)
 
         if st.session_state.get("__enable_video"):
-            st.file_uploader(
-                """
-                ##### 👩‍🦰 Input Face
-                Upload a video/image that contains faces to use
-                *Recommended - mp4 / mov / png / jpg / gif*
-                """,
-                key="input_face",
-            )
             lipsync_settings()
 
         if st.session_state.get("__enable_translation"):
-            google_translate_language_selector(
-                f"##### {field_title_desc(self.RequestModel, 'user_language')}",
-                key="user_language",
-            )
+            st.markdown("##### 🔠 Translation Settings")
             enable_glossary = st.checkbox(
                 "📖 Add Glossary",
                 value=bool(