mrq/ai-voice-cloning
31
74
Fork 43
Code Issues 223 Pull Requests 5 Packages Projects Releases Wiki Activity

added parameter to specify the autoregressive_model (tho it still loads the default model first, and then loads the target model, which seems to be because TTS loading just loads whatever is set in the settings first)

Browse Source 
removed load_tts() call, the cli tool now relys on generate() to load the tts system, which is less fail prone imho
This commit is contained in:
ben_mkiv 2023-08-22 23:33:02 +02:00
parent 89102347a9
commit d1dbe3e464
1 changed files with 4 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
src/cli.py
View File

@ -32,6 +32,7 @@ if __name__ == "__main__":
parser.add_argument("--breathing_room", default=default_arguments['breathing_room']) parser.add_argument("--breathing_room", default=default_arguments['breathing_room'])
parser.add_argument("--cvvp_weight", default=default_arguments['cvvp_weight']) parser.add_argument("--cvvp_weight", default=default_arguments['cvvp_weight'])
parser.add_argument("--top_p", default=default_arguments['top_p']) parser.add_argument("--top_p", default=default_arguments['top_p'])
parser.add_argument("--autoregressive_model", default=default_arguments['autoregressive_model'])
parser.add_argument("--diffusion_temperature", default=default_arguments['diffusion_temperature']) parser.add_argument("--diffusion_temperature", default=default_arguments['diffusion_temperature'])
parser.add_argument("--length_penalty", default=default_arguments['length_penalty']) parser.add_argument("--length_penalty", default=default_arguments['length_penalty'])
parser.add_argument("--repetition_penalty", default=default_arguments['repetition_penalty']) parser.add_argument("--repetition_penalty", default=default_arguments['repetition_penalty'])
@ -55,6 +56,7 @@ if __name__ == "__main__":
'breathing_room': args.breathing_room, 'breathing_room': args.breathing_room,
'cvvp_weight': args.cvvp_weight, 'cvvp_weight': args.cvvp_weight,
'top_p': args.top_p, 'top_p': args.top_p,
'autoregressive_model': args.autoregressive_model,
'diffusion_temperature': args.diffusion_temperature, 'diffusion_temperature': args.diffusion_temperature,
'length_penalty': args.length_penalty, 'length_penalty': args.length_penalty,
'repetition_penalty': args.repetition_penalty, 'repetition_penalty': args.repetition_penalty,
@ -62,5 +64,6 @@ if __name__ == "__main__":
'experimentals': default_arguments['experimentals'], 'experimentals': default_arguments['experimentals'],
} }
tts = load_tts() # cli should rely on generate() for loading the TTS backend
#tts = load_tts()
generate(**kwargs) generate(**kwargs)