Cantonese TTS MFA pipeline, test=tts

pull/2899/head
WongLaw 3 years ago
parent d9df9b05ca
commit cfa5a5021f

@ -32,12 +32,14 @@ pyworld>=0.2.12
recommonmark>=0.5.0 recommonmark>=0.5.0
resampy resampy
sacrebleu sacrebleu
shutil
sphinx sphinx
sphinx-autobuild sphinx-autobuild
sphinx-markdown-tables sphinx-markdown-tables
sphinx_rtd_theme sphinx_rtd_theme
textgrid textgrid
timer timer
ToJyutping
typeguard typeguard
webrtcvad webrtcvad
websockets websockets

@ -9,7 +9,7 @@ If you want to get rhythm tags with duration through MFA tool, you may add flag
Note that only CSMSC dataset is supported so far, and we replace `#` with `sp` in rhythm tags for MFA. Note that only CSMSC dataset is supported so far, and we replace `#` with `sp` in rhythm tags for MFA.
# MFA for Cantonese language # MFA for Cantonese language
First, go download these datasets [Guangzhou_Cantonese_Scripted_Speech_Corpus_Daily_Use_Sentence](https://paddlespeech.bj.bcebos.com/datasets/Cantonese/Guangzhou_Cantonese_Scripted_Speech_Corpus_Daily_Use_Sentence.zip) and [Guangzhou_Cantonese_Scripted_Speech_Corpus_in_Vehicle](https://paddlespeech.bj.bcebos.com/datasets/Cantonese/Guangzhou_Cantonese_Scripted_Speech_Corpus_in_Vehicle.zip) under `~/datasets/`. First, go download these datasets [Guangzhou_Cantonese_Scripted_Speech_Corpus_Daily_Use_Sentence](https://magichub.com/datasets/guangzhou-cantonese-scripted-speech-corpus-daily-use-sentence/) and [Guangzhou_Cantonese_Scripted_Speech_Corpus_in_Vehicle](https://magichub.com/datasets/guangzhou-cantonese-scripted-speech-corpus-in-the-vehicle/) under `~/datasets/`.
Then, Then,
```bash ```bash
./run_canton.sh ./run_canton.sh

@ -1,3 +1,4 @@
import argparse
import os import os
import re import re
import shutil import shutil
@ -34,7 +35,9 @@ if __name__ == "__main__":
parser.add_argument( parser.add_argument(
"--output_lexicon", type=str, help="Path to save lexicon.") "--output_lexicon", type=str, help="Path to save lexicon.")
parser.add_argument( parser.add_argument(
"--output_wavlabs", type=str, help="Path to save lexicon.") "--output_wavlabs",
type=str,
help="Path of wavs and labs for MFA training.")
parser.add_argument( parser.add_argument(
"--inputs", "--inputs",
type=str, type=str,

@ -67,6 +67,7 @@ base = [
"pyyaml", "pyyaml",
"paddleslim>=2.3.4", "paddleslim>=2.3.4",
"paddleaudio>=1.1.0", "paddleaudio>=1.1.0",
"ToJyutping",
] ]
server = ["pattern_singleton", "websockets"] server = ["pattern_singleton", "websockets"]
@ -292,7 +293,8 @@ setup_info = dict(
}, },
# Package info # Package info
packages=find_packages(include=['paddlespeech*'], exclude=['utils', 'third_party']), packages=find_packages(
include=['paddlespeech*'], exclude=['utils', 'third_party']),
zip_safe=True, zip_safe=True,
classifiers=[ classifiers=[
'Development Status :: 5 - Production/Stable', 'Development Status :: 5 - Production/Stable',

Loading…
Cancel
Save