blackteay/GPT-SoVITS
1
0
Fork 0
mirror of https://github.com/RVC-Boss/GPT-SoVITS.git synced 2026-06-03 20:40:30 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge 8a70a2ffd0c0b81af31c912b5dcce535dcdb6268 into 08d627c3338173c3229286d8787060d6559fe0f8

Browse Source
This commit is contained in:
DayDaySpeed 2026-06-01 20:02:09 +08:00 committed by GitHub
commit 5a8f5cbd36
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 34 additions and 30 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

64
tools/asr/funasr_asr.py
View File

@ -22,51 +22,55 @@ def only_asr(input_file, language):
def create_model(language="zh"): def create_model(language="zh"):
if language in funasr_models:
return funasr_models[language]
if language == "zh": if language == "zh":
path_vad = "tools/asr/models/speech_fsmn_vad_zh-cn-16k-common-pytorch" path_vad = "tools/asr/models/speech_fsmn_vad_zh-cn-16k-common-pytorch"
path_punc = "tools/asr/models/punc_ct-transformer_zh-cn-common-vocab272727-pytorch" path_punc = "tools/asr/models/punc_ct-transformer_zh-cn-common-vocab272727-pytorch"
path_asr = "tools/asr/models/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch" path_asr = "tools/asr/models/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
snapshot_download( if not os.path.isdir(path_vad):
"iic/speech_fsmn_vad_zh-cn-16k-common-pytorch", snapshot_download(
local_dir="tools/asr/models/speech_fsmn_vad_zh-cn-16k-common-pytorch", "iic/speech_fsmn_vad_zh-cn-16k-common-pytorch",
) local_dir=path_vad,
snapshot_download( )
"iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch", if not os.path.isdir(path_punc):
local_dir="tools/asr/models/punc_ct-transformer_zh-cn-common-vocab272727-pytorch", snapshot_download(
) "iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch",
snapshot_download( local_dir=path_punc,
"iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch", )
local_dir="tools/asr/models/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch", if not os.path.isdir(path_asr):
) snapshot_download(
"iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch",
local_dir=path_asr,
)
model_revision = "v2.0.4" model_revision = "v2.0.4"
vad_model_revision = punc_model_revision = "v2.0.4" vad_model_revision = punc_model_revision = "v2.0.4"
elif language == "yue": elif language == "yue":
path_asr = "tools/asr/models/speech_UniASR_asr_2pass-cantonese-CHS-16k-common-vocab1468-tensorflow1-online" path_asr = "tools/asr/models/speech_UniASR_asr_2pass-cantonese-CHS-16k-common-vocab1468-tensorflow1-online"
snapshot_download( if not os.path.isdir(path_asr):
"iic/speech_UniASR_asr_2pass-cantonese-CHS-16k-common-vocab1468-tensorflow1-online", snapshot_download(
local_dir="tools/asr/models/speech_UniASR_asr_2pass-cantonese-CHS-16k-common-vocab1468-tensorflow1-online", "iic/speech_UniASR_asr_2pass-cantonese-CHS-16k-common-vocab1468-tensorflow1-online",
) local_dir=path_asr,
)
path_vad = path_punc = None path_vad = path_punc = None
vad_model_revision = punc_model_revision = "" vad_model_revision = punc_model_revision = ""
model_revision = "master" model_revision = "master"
else: else:
raise ValueError(f"{language} is not supported") raise ValueError(f"{language} is not supported")
if language in funasr_models: model = AutoModel(
return funasr_models[language] model=path_asr,
else: model_revision=model_revision,
model = AutoModel( vad_model=path_vad,
model=path_asr, vad_model_revision=vad_model_revision,
model_revision=model_revision, punc_model=path_punc,
vad_model=path_vad, punc_model_revision=punc_model_revision,
vad_model_revision=vad_model_revision, )
punc_model=path_punc, print(f"FunASR 模型加载完成: {language.upper()}")
punc_model_revision=punc_model_revision,
)
print(f"FunASR 模型加载完成: {language.upper()}")
funasr_models[language] = model funasr_models[language] = model
return model return model
def execute_asr(input_folder, output_folder, model_size, language): def execute_asr(input_folder, output_folder, model_size, language):