dianskip/ai-voice-cloning
1
0
Fork 0
forked from camenduru/ai-voice-cloning
Code Pull Requests Activity

fixed the whisperx has no attribute named load_model whatever because I guess whisperx has as stable of an API as I do

Browse Source
This commit is contained in:
mrq 2023-05-06 10:45:17 +00:00
parent e416b0fe6f
commit 149aaca554
1 changed files with 5 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
src/utils.py
View File

@ -3660,9 +3660,12 @@ def load_whisper_model(language=None, model_name=None, progress=None):
b_lang = language.encode('ascii')
whisper_model = Whisper(model_name, models_dir='./models/', language=b_lang)
elif args.whisper_backend == "m-bain/whisperx":
import whisperx
import whisper, whisperx
device = "cuda" if get_device_name() == "cuda" else "cpu"
whisper_model = whisperx.load_model(model_name, device)
try:
whisper_model = whisperx.load_model(model_name, device)
except Exception as e:
whisper_model = whisper.load_model(model_name, device)
if not args.hf_token:
print("No huggingface token used, needs to be saved in environment variable, otherwise will throw error loading VAD model.")