Skip to content

Commit

Permalink
Add more samples
Browse files Browse the repository at this point in the history
synesthesiam committed May 30, 2024

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
1 parent ebb6d50 commit 521a454
Showing 13 changed files with 130 additions and 4 deletions.
18 changes: 17 additions & 1 deletion _script/generate-samples.sh
Original file line number Diff line number Diff line change
@@ -42,6 +42,21 @@ find "${piper_voices}" -name '*.onnx' | sort | \
samples_dir="${voice_dir}/samples"
mkdir -p "${samples_dir}"

noise_scale="$(jq --raw-output '.inference.noise_scale' "${onnx}.json")"
if [ -z "${noise_scale}" ]; then
noise_scale="0.667"
fi

length_scale="$(jq --raw-output '.inference.length_scale' "${onnx}.json")"
if [ -z "${length_scale}" ]; then
length_scale="1.0"
fi

noise_w="$(jq --raw-output '.inference.noise_w' "${onnx}.json")"
if [ -z "${noise_w}" ]; then
noise_w="0.8"
fi

num_speakers="$(jq --raw-output '.num_speakers' "${onnx}.json")"
sample_rate="$(jq --raw-output '.audio.sample_rate' "${onnx}.json")"
last_speaker_id="$((num_speakers-1))"
@@ -60,7 +75,8 @@ find "${piper_voices}" -name '*.onnx' | sort | \

# Compress to MP3 with ffmpeg
head -n1 "${test_sentences}" | \
"${piper_binary}" --model "${onnx}" --speaker "${speaker_id}" --output_raw | \
"${piper_binary}" --model "${onnx}" --speaker "${speaker_id}" --output_raw \
--noise_scale "${noise_scale}" --noise_w "${noise_w}" --length_scale "${length_scale}" | \
ffmpeg -hide_banner -loglevel warning -y \
-sample_rate "${sample_rate}" -f s16le -ac 1 -i - \
-codec:a libmp3lame -qscale:a 2 "${sample_mp3}";
15 changes: 15 additions & 0 deletions samples/cy/cy_GB/gwryw_gogleddol/medium/MODEL_CARD
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
# Model card for gwryw_gogleddol (medium)

* Language: cy_GB (Welsh, Great Britain)
* Speakers: 1
* Quality: medium
* Samplerate: 22,050Hz

## Dataset

* URL: https://git.techiaith.bangor.ac.uk/data-porth-technolegau-iaith/corpws-talentau-llais
* License: See URL

## Training

Finetuned from U.S. English lessac voice (medium quality).
1 change: 1 addition & 0 deletions samples/cy/cy_GB/gwryw_gogleddol/medium/sample.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
Rhyfeddod neu ffenomenon optegol a meteorolegol yw enfys, pan fydd sbectrwm o olau yn ymddangos yn yr awyr pan fo'r haul yn disgleirio ar ddiferion o leithder yn atmosffer y ddaear.
Binary file not shown.
2 changes: 1 addition & 1 deletion samples/en/en_GB/cori/high/MODEL_CARD
Original file line number Diff line number Diff line change
@@ -2,7 +2,7 @@

* Language: en_GB (English, Great Britain)
* Speakers: 1
* Quality: medium
* Quality: high
* Samplerate: 22,050Hz

## Dataset
17 changes: 17 additions & 0 deletions samples/en/en_GB/cori/medium/MODEL_CARD
Original file line number Diff line number Diff line change
@@ -0,0 +1,17 @@
# Model card for cori (medium)

* Language: en_GB (English, Great Britain)
* Speakers: 1
* Quality: medium
* Samplerate: 22,050Hz

## Dataset

* URL: https://librivox.org
* License: public domain

## Training

See: https://brycebeattie.com/files/tts/

UK English female voice. Single Speaker. Trained from scratch on medium quality settings for 640 epochs. I put together the dataset, which ended up with about 24 hours of recordings. All recordings came from LibriVox.org.
1 change: 1 addition & 0 deletions samples/en/en_GB/cori/medium/sample.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
A rainbow is a meteorological phenomenon that is caused by reflection, refraction and dispersion of light in water droplets resulting in a spectrum of light appearing in the sky.
Binary file added samples/en/en_GB/cori/medium/speaker_0.mp3
Binary file not shown.
Binary file modified samples/fa/fa_IR/gyro/medium/speaker_0.mp3
Binary file not shown.
15 changes: 15 additions & 0 deletions samples/it/it_IT/paola/medium/MODEL_CARD
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
# Model card for paola (medium)

* Language: it_IT (Italian, Italy)
* Speakers: 1
* Quality: medium
* Samplerate: 22,050Hz

## Dataset

* URL: https://huggingface.co/datasets/paolapersico1/Voice-Dataset-Italian
* License: See URL

## Training

Finetuned from U.S. English lessac voice (medium quality).
1 change: 1 addition & 0 deletions samples/it/it_IT/paola/medium/sample.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
In fisica dell'atmosfera e meteorologia l'arcobaleno è un fenomeno ottico atmosferico che produce uno spettro quasi continuo di luce nel cielo quando la luce del Sole attraversa le gocce d'acqua rimaste in sospensione dopo un temporale, o presso una cascata o una fontana.
Binary file added samples/it/it_IT/paola/medium/speaker_0.mp3
Binary file not shown.
64 changes: 62 additions & 2 deletions voices.json
Original file line number Diff line number Diff line change
@@ -213,6 +213,36 @@
},
"aliases": []
},
"cy_GB-gwryw_gogleddol-medium": {
"key": "cy_GB-gwryw_gogleddol-medium",
"name": "gwryw_gogleddol",
"language": {
"code": "cy_GB",
"family": "cy",
"region": "GB",
"name_native": "Cymraeg",
"name_english": "Welsh",
"country_english": "Great Britain"
},
"quality": "medium",
"num_speakers": 1,
"speaker_id_map": {},
"files": {
"cy/cy_GB/gwryw_gogleddol/medium/cy_GB-gwryw_gogleddol-medium.onnx": {
"size_bytes": 63511038,
"md5_digest": "76ca79c170b0048b190758c3609e9ab9"
},
"cy/cy_GB/gwryw_gogleddol/medium/cy_GB-gwryw_gogleddol-medium.onnx.json": {
"size_bytes": 4975,
"md5_digest": "d780e83a324e9ce8c73146b9d066a283"
},
"cy/cy_GB/gwryw_gogleddol/medium/MODEL_CARD": {
"size_bytes": 337,
"md5_digest": "39bd3ade08289afe1cb3f867aef957fa"
}
},
"aliases": []
},
"da_DK-talesyntese-medium": {
"key": "da_DK-talesyntese-medium",
"name": "talesyntese",
@@ -823,8 +853,8 @@
"md5_digest": "04e0151b653bb64540b1cde027054140"
},
"el/el_GR/rapunzelina/low/el_GR-rapunzelina-low.onnx.json": {
"size_bytes": 4169,
"md5_digest": "b2b064991a72745a5b444af536afbe86"
"size_bytes": 4198,
"md5_digest": "8d6cd8a576008116be5281b13e1c7b45"
},
"el/el_GR/rapunzelina/low/MODEL_CARD": {
"size_bytes": 303,
@@ -4701,6 +4731,36 @@
"is-ugla-medium"
]
},
"it_IT-paola-medium": {
"key": "it_IT-paola-medium",
"name": "paola",
"language": {
"code": "it_IT",
"family": "it",
"region": "IT",
"name_native": "Italiano",
"name_english": "Italian",
"country_english": "Italy"
},
"quality": "medium",
"num_speakers": 1,
"speaker_id_map": {},
"files": {
"it/it_IT/paola/medium/it_IT-paola-medium.onnx": {
"size_bytes": 63511038,
"md5_digest": "3a44e73b12ca5d0c21a72e388b5847c8"
},
"it/it_IT/paola/medium/it_IT-paola-medium.onnx.json": {
"size_bytes": 7100,
"md5_digest": "3c983f7292518f042b4d137b2dfe65b3"
},
"it/it_IT/paola/medium/MODEL_CARD": {
"size_bytes": 303,
"md5_digest": "436971e8acb0a92dd8dbc42542e59d03"
}
},
"aliases": []
},
"it_IT-riccardo-x_low": {
"key": "it_IT-riccardo-x_low",
"name": "riccardo",

0 comments on commit 521a454

Please sign in to comment.