blackteay/GPT-SoVITS
1
0
Fork 0
mirror of https://github.com/RVC-Boss/GPT-SoVITS.git synced 2025-10-07 07:14:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

add punc model

Browse Source
This commit is contained in:
XXXXRT666 2024-07-13 23:00:24 +08:00
parent 9937b78e1a
commit bae36bb07a
1 changed files with 11 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
tools/asr/sensevoice_asr.py
View File

@ -6,14 +6,19 @@ import torch
from tqdm import tqdm from tqdm import tqdm
from funasr import AutoModel from funasr import AutoModel
model_dir = "tools/asr/models/SenseVoiceSmall" path_asr = "tools/asr/models/SenseVoiceSmall"
model_dir = model_dir if os.path.exists(model_dir) else "iic/SenseVoiceSmall" 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_asr = path_asr if os.path.exists(path_asr) else "iic/SenseVoiceSmall"
path_vad = path_vad if os.path.exists(path_vad) else "iic/speech_fsmn_vad_zh-cn-16k-common-pytorch"
path_punc = path_punc if os.path.exists(path_punc) else "iic/punc_ct-transformer_zh-cn-common-vocab272727-pytorch"
def execute_asr(input_folder, output_folder, language, device): def execute_asr(input_folder, output_folder, language, device):
try: try:
model = AutoModel(model=model_dir, model = AutoModel(model=path_asr,
vad_model="fsmn-vad", vad_model=path_vad,
vad_kwargs={"max_single_segment_time": 30000}, vad_kwargs={"max_single_segment_time": 2000},
punc_model=path_punc
) )
except: except:
return print(traceback.format_exc()) return print(traceback.format_exc())
@ -31,7 +36,7 @@ def execute_asr(input_folder, output_folder, language, device):
input=file_path, input=file_path,
cache={}, cache={},
language=language.lower(), # "zn", "en", "yue", "ja", "ko", "nospeech" language=language.lower(), # "zn", "en", "yue", "ja", "ko", "nospeech"
use_itn=True, use_itn=False,
batch_size_s=0, batch_size_s=0,
device = device device = device
)[0]['text'] )[0]['text']