Skip to content
This repository has been archived by the owner on Jun 21, 2023. It is now read-only.

Commit

Permalink
Add samples page
Browse files Browse the repository at this point in the history
  • Loading branch information
Michael Hansen committed Mar 26, 2021
1 parent 13311e2 commit 8f3b3e0
Show file tree
Hide file tree
Showing 346 changed files with 5,173 additions and 0 deletions.
678 changes: 678 additions & 0 deletions index.html

Large diffs are not rendered by default.

5 changes: 5 additions & 0 deletions local/de-de/thorsten-glow_tts/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
# German Larynx Voice (thorsten)

[GlowTTS](https://github.com/rhasspy/glow-tts-train) model trained from the [thorsten dataset](https://github.com/thorstenMueller/deep-learning-german-tts/).

Intended to be used with [gruut](https://github.com/rhasspy/gruut) (`de-de`).
57 changes: 57 additions & 0 deletions local/de-de/thorsten-glow_tts/config.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,57 @@
{
"seed": 1234,
"epochs": 10000,
"learning_rate": 0.00026855278715645745,
"betas": [
0.9,
0.98
],
"eps": 1e-09,
"grad_clip": 5.0,
"warmup_steps": 4000,
"scheduler": "noam",
"batch_size": 32,
"fp16_run": false,
"audio": {
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"mel_channels": 80,
"sample_rate": 22050,
"sample_bytes": 2,
"channels": 1,
"mel_fmin": 0.0,
"mel_fmax": 8000.0,
"signal_norm": false,
"convert_db_to_amp": false,
"do_dynamic_range_compression": false
},
"model": {
"num_symbols": 54,
"hidden_channels": 192,
"filter_channels": 768,
"filter_channels_dp": 256,
"kernel_size": 3,
"p_dropout": 0.1,
"n_blocks_dec": 12,
"n_layers_enc": 6,
"n_heads": 2,
"p_dropout_dec": 0.05,
"dilation_rate": 1,
"kernel_size_dec": 5,
"n_block_layers": 4,
"n_sqz": 2,
"prenet": true,
"mean_only": true,
"hidden_channels_enc": 192,
"hidden_channels_dec": 192,
"window_size": 4,
"n_speakers": 1,
"n_split": 4,
"sigmoid_scale": false,
"block_length": null,
"gin_channels": 0
},
"version": 1,
"git_commit": "a303dc3"
}
54 changes: 54 additions & 0 deletions local/de-de/thorsten-glow_tts/phonemes.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1,54 @@
0 _
1 |
2 ‖
3 #
4 a
5 aɪ̯
6 aʊ̯
7 aː
8 b
9 d
10 d͡ʒ
11 eː
12 f
13 g
14 h
15 iː
16 j
17 k
18 l
19 m
20 n
21 oː
22 p
23 p͡f
24 s
25 t
26 t͡s
27 t͡ʃ
28 uː
29 v
30 x
31 yː
32 z
33 ãː
34 ç
35 õː
36 øː
37 ŋ
38 œ
39 ɐ
40 ɔ
41 ɔʏ̯
42 ə
43 ɛ
44 ɛː
45 ɛ̃ː
46 ɪ
47 ʁ
48 ʃ
49 ʊ
50 ʏ
51 ʒ
52 ʔ
53 χ
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
5 changes: 5 additions & 0 deletions local/de-de/thorsten-glow_tts/samples/test_phonemes.csv
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
konnen_sie_bitte|3 17 38 20 42 20 3 32 15 3 8 46 25 42 3 18 4 37 32 7 19 39 3 48 22 47 43 34 20 3 2
mir_geht_es_gut|3 19 15 39 3 13 11 25 3 52 11 24 3 13 28 25 3 1 3 9 4 37 17 42 3 2
haben_sie_ein_vegetarisches|3 14 7 8 42 20 3 32 15 3 5 20 3 29 11 13 11 25 4 47 46 48 42 24 3 13 42 47 46 34 25 3 2
ich_bin_allergisch|3 46 34 3 8 46 20 3 52 4 18 18 43 47 13 46 48 3 2
fischers_fritze_fischt|3 12 46 48 43 47 24 3 12 47 46 25 24 42 3 12 46 48 25 3 12 47 46 48 42 3 12 46 48 42 3 1 3 12 47 46 48 42 3 12 46 48 42 3 12 46 48 25 3 12 46 48 43 47 24 3 12 47 46 25 24 42 3 2
5 changes: 5 additions & 0 deletions local/de-de/thorsten-glow_tts/samples/test_sentences.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
konnen_sie_bitte|Können Sie bitte langsamer sprechen?
mir_geht_es_gut|Mir geht es gut, danke!
haben_sie_ein_vegetarisches|Haben Sie ein vegetarisches Gericht?
ich_bin_allergisch|Ich bin allergisch.
fischers_fritze_fischt|Fischers Fritze fischt frische Fische; Frische Fische fischt Fischers Fritze.
5 changes: 5 additions & 0 deletions local/en-us/blizzard_fls-glow_tts/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
# English Larynx Voice (blizzard_fls)

[GlowTTS](https://github.com/rhasspy/glow-tts-train) model trained from the [Blizzard 2017 Challenge dataset](https://www.synsig.org/index.php/Blizzard_Challenge_2017).

Intended to be used with [gruut](https://github.com/rhasspy/gruut) (`en-us`).
64 changes: 64 additions & 0 deletions local/en-us/blizzard_fls-glow_tts/config.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,64 @@
{
"seed": 1234,
"epochs": 10000,
"learning_rate": 0.0005261910343112216,
"betas": [
0.9,
0.98
],
"eps": 1e-09,
"grad_clip": 5.0,
"warmup_steps": 4000,
"scheduler": "noam",
"batch_size": 32,
"fp16_run": false,
"min_seq_length": null,
"max_seq_length": null,
"audio": {
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"mel_channels": 80,
"sample_rate": 22050,
"sample_bytes": 2,
"channels": 1,
"mel_fmin": 0.0,
"mel_fmax": 8000.0,
"ref_level_db": 20.0,
"spec_gain": 1.0,
"signal_norm": true,
"min_level_db": -100.0,
"max_norm": 1.0,
"clip_norm": true,
"symmetric_norm": true
},
"model": {
"num_symbols": 46,
"hidden_channels": 192,
"filter_channels": 768,
"filter_channels_dp": 256,
"kernel_size": 3,
"p_dropout": 0.1,
"n_blocks_dec": 12,
"n_layers_enc": 6,
"n_heads": 2,
"p_dropout_dec": 0.05,
"dilation_rate": 1,
"kernel_size_dec": 5,
"n_block_layers": 4,
"n_sqz": 2,
"prenet": true,
"mean_only": true,
"hidden_channels_enc": 192,
"hidden_channels_dec": 192,
"window_size": 4,
"n_speakers": 1,
"n_split": 4,
"sigmoid_scale": false,
"block_length": null,
"gin_channels": 0,
"n_frames_per_step": 1
},
"version": 1,
"git_commit": "b05e772"
}
46 changes: 46 additions & 0 deletions local/en-us/blizzard_fls-glow_tts/phonemes.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1,46 @@
0 _
1 |
2 ‖
3 #
4 ˈ
5 ˌ
6 aɪ
7 aʊ
8 b
9 d
10 d͡ʒ
11 eɪ
12 f
13 h
14 i
15 j
16 k
17 l
18 m
19 n
20 oʊ
21 p
22 s
23 t
24 t͡ʃ
25 u
26 v
27 w
28 z
29 æ
30 ð
31 ŋ
32 ɑ
33 ɔ
34 ɔɪ
35 ə
36 ɚ
37 ɛ
38 ɡ
39 ɪ
40 ɹ
41 ʃ
42 ʊ
43 ʌ
44 ʒ
45 θ
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
5 changes: 5 additions & 0 deletions local/en-us/blizzard_fls-glow_tts/samples/test_phonemes.csv
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
it_took_me_quite_a_long_time_to_develop_a_voice|3 4 39 23 3 23 4 42 16 3 18 4 14 3 16 27 4 6 23 3 35 3 17 4 33 31 3 23 4 6 18 3 23 4 25 3 9 39 26 4 37 17 35 21 3 35 3 26 4 34 22 3 1 3 35 19 9 3 19 4 7 3 30 4 29 23 3 4 6 3 13 4 29 26 3 4 39 23 3 4 6 18 3 19 4 32 23 3 38 4 20 39 31 3 23 4 25 3 8 4 14 3 22 4 6 17 35 19 23 3 2
be_a_voice_not_an_echo|3 8 4 14 3 35 3 26 4 34 22 3 1 3 19 4 32 23 3 35 19 3 4 37 16 20 3 2
im_sorry_dave|3 4 6 18 3 22 4 32 40 14 3 9 4 11 26 3 2 4 6 18 3 35 12 40 4 11 9 3 4 6 3 16 4 29 19 23 3 9 4 25 3 30 4 29 23 3 2
this_cake_is_great|3 30 4 39 22 3 16 4 11 16 3 4 39 28 3 38 40 4 11 23 3 2 4 39 23 22 3 22 4 20 3 9 39 17 4 39 41 35 22 3 35 19 9 3 18 4 34 22 23 3 2
prior_to_november|3 21 40 4 6 36 3 23 4 25 3 19 20 26 4 37 18 8 36 3 23 27 4 37 19 23 14 3 22 4 37 16 35 19 9 3 1 3 19 4 6 19 23 4 14 19 3 22 4 39 16 22 23 14 3 45 40 4 14 3 2
5 changes: 5 additions & 0 deletions local/en-us/blizzard_fls-glow_tts/samples/test_sentences.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
it_took_me_quite_a_long_time_to_develop_a_voice|It took me quite a long time to develop a voice, and now that I have it I'm not going to be silent.
be_a_voice_not_an_echo|Be a voice, not an echo.
im_sorry_dave|I'm sorry Dave. I'm afraid I can't do that.
this_cake_is_great|This cake is great. It's so delicious and moist.
prior_to_november|Prior to November 22_ordinal, 1963_year.
Binary file not shown.
5 changes: 5 additions & 0 deletions local/en-us/cmu_aew-glow_tts/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
# English Larynx Voice (aew)

[GlowTTS](https://github.com/rhasspy/glow-tts-train) model trained from the [CMU Arctic](http://www.festvox.org/cmu_arctic/).

Intended to be used with [gruut](https://github.com/rhasspy/gruut) (`en-us`).
64 changes: 64 additions & 0 deletions local/en-us/cmu_aew-glow_tts/config.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,64 @@
{
"seed": 1234,
"epochs": 1000,
"learning_rate": 0.0003184746968987751,
"betas": [
0.9,
0.98
],
"eps": 1e-09,
"grad_clip": 5.0,
"warmup_steps": 4000,
"scheduler": "noam",
"batch_size": 32,
"fp16_run": false,
"min_seq_length": null,
"max_seq_length": 153,
"audio": {
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"mel_channels": 80,
"sample_rate": 22050,
"sample_bytes": 2,
"channels": 1,
"mel_fmin": 0.0,
"mel_fmax": 8000.0,
"ref_level_db": 20.0,
"spec_gain": 1.0,
"signal_norm": true,
"min_level_db": -100.0,
"max_norm": 1.0,
"clip_norm": true,
"symmetric_norm": true
},
"model": {
"num_symbols": 46,
"hidden_channels": 192,
"filter_channels": 768,
"filter_channels_dp": 256,
"kernel_size": 3,
"p_dropout": 0.1,
"n_blocks_dec": 12,
"n_layers_enc": 6,
"n_heads": 2,
"p_dropout_dec": 0.05,
"dilation_rate": 1,
"kernel_size_dec": 5,
"n_block_layers": 4,
"n_sqz": 2,
"prenet": true,
"mean_only": true,
"hidden_channels_enc": 192,
"hidden_channels_dec": 192,
"window_size": 4,
"n_speakers": 1,
"n_split": 4,
"sigmoid_scale": false,
"block_length": null,
"gin_channels": 0,
"n_frames_per_step": 1
},
"version": 1,
"git_commit": "b05e772"
}
46 changes: 46 additions & 0 deletions local/en-us/cmu_aew-glow_tts/phonemes.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1,46 @@
0 _
1 |
2 ‖
3 #
4 ˈ
5 ˌ
6 aɪ
7 aʊ
8 b
9 d
10 d͡ʒ
11 eɪ
12 f
13 h
14 i
15 j
16 k
17 l
18 m
19 n
20 oʊ
21 p
22 s
23 t
24 t͡ʃ
25 u
26 v
27 w
28 z
29 æ
30 ð
31 ŋ
32 ɑ
33 ɔ
34 ɔɪ
35 ə
36 ɚ
37 ɛ
38 ɡ
39 ɪ
40 ɹ
41 ʃ
42 ʊ
43 ʌ
44 ʒ
45 θ
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
5 changes: 5 additions & 0 deletions local/en-us/cmu_aew-glow_tts/samples/test_phonemes.csv
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
it_took_me_quite_a_long_time_to_develop_a_voice|3 4 39 23 3 23 4 42 16 3 18 4 14 3 16 27 4 6 23 3 35 3 17 4 33 31 3 23 4 6 18 3 23 4 25 3 9 39 26 4 37 17 35 21 3 35 3 26 4 34 22 3 1 3 35 19 9 3 19 4 7 3 30 4 29 23 3 4 6 3 13 4 29 26 3 4 39 23 3 4 6 18 3 19 4 32 23 3 38 4 20 39 31 3 23 4 25 3 8 4 14 3 22 4 6 17 35 19 23 3 2
be_a_voice_not_an_echo|3 8 4 14 3 35 3 26 4 34 22 3 1 3 19 4 32 23 3 35 19 3 4 37 16 20 3 2
im_sorry_dave|3 4 6 18 3 22 4 32 40 14 3 9 4 11 26 3 2 4 6 18 3 35 12 40 4 11 9 3 4 6 3 16 4 29 19 23 3 9 4 25 3 30 4 29 23 3 2
this_cake_is_great|3 30 4 39 22 3 16 4 11 16 3 4 39 28 3 38 40 4 11 23 3 2 4 39 23 22 3 22 4 20 3 9 39 17 4 39 41 35 22 3 35 19 9 3 18 4 34 22 23 3 2
prior_to_november|3 21 40 4 6 36 3 23 4 25 3 19 20 26 4 37 18 8 36 3 23 27 4 37 19 23 14 3 22 4 37 16 35 19 9 3 1 3 19 4 6 19 23 4 14 19 3 22 4 39 16 22 23 14 3 45 40 4 14 3 2
5 changes: 5 additions & 0 deletions local/en-us/cmu_aew-glow_tts/samples/test_sentences.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
it_took_me_quite_a_long_time_to_develop_a_voice|It took me quite a long time to develop a voice, and now that I have it I'm not going to be silent.
be_a_voice_not_an_echo|Be a voice, not an echo.
im_sorry_dave|I'm sorry Dave. I'm afraid I can't do that.
this_cake_is_great|This cake is great. It's so delicious and moist.
prior_to_november|Prior to November 22_ordinal, 1963_year.
Binary file not shown.
5 changes: 5 additions & 0 deletions local/en-us/cmu_ahw-glow_tts/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
# English Larynx Voice (ahw)

[GlowTTS](https://github.com/rhasspy/glow-tts-train) model trained from the [CMU Arctic](http://www.festvox.org/cmu_arctic/).

Intended to be used with [gruut](https://github.com/rhasspy/gruut) (`en-us`).
Loading

0 comments on commit 8f3b3e0

Please sign in to comment.