psr/ai-voice-cloning
1
0
Fork 0
forked from mrq/ai-voice-cloning
Code Issues Pull Requests Packages Projects Releases Wiki Activity

option to set results folder location

Browse Source
This commit is contained in:
mrq 2023-03-19 22:03:41 +00:00
parent 74510e8623
commit 5cb86106ce
2 changed files with 12 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
src/utils.py
View File

@ -252,7 +252,7 @@ def generate(**kwargs):
full_start_time = time.time() full_start_time = time.time()
outdir = f"./results/{voice}/" outdir = f"./{args.results_folder}/{voice}/"
os.makedirs(outdir, exist_ok=True) os.makedirs(outdir, exist_ok=True)
audio_cache = {} audio_cache = {}
@ -2024,6 +2024,7 @@ def setup_args():
'output-sample-rate': 44100, 'output-sample-rate': 44100,
'output-volume': 1, 'output-volume': 1,
'results-folder': "./results/",
'tts-backend': TTSES[0], 'tts-backend': TTSES[0],
@ -2070,6 +2071,7 @@ def setup_args():
parser.add_argument("--autocalculate-voice-chunk-duration-size", type=float, default=default_arguments['autocalculate-voice-chunk-duration-size'], help="Number of seconds to suggest voice chunk size for (for example, 100 seconds of audio at 10 seconds per chunk will suggest 10 chunks)") parser.add_argument("--autocalculate-voice-chunk-duration-size", type=float, default=default_arguments['autocalculate-voice-chunk-duration-size'], help="Number of seconds to suggest voice chunk size for (for example, 100 seconds of audio at 10 seconds per chunk will suggest 10 chunks)")
parser.add_argument("--output-sample-rate", type=int, default=default_arguments['output-sample-rate'], help="Sample rate to resample the output to (from 24KHz)") parser.add_argument("--output-sample-rate", type=int, default=default_arguments['output-sample-rate'], help="Sample rate to resample the output to (from 24KHz)")
parser.add_argument("--output-volume", type=float, default=default_arguments['output-volume'], help="Adjusts volume of output") parser.add_argument("--output-volume", type=float, default=default_arguments['output-volume'], help="Adjusts volume of output")
parser.add_argument("--results-folder", type=str, default=default_arguments['results-folder'], help="Sets output directory")
parser.add_argument("--tts-backend", default=default_arguments['tts-backend'], help="Specifies which TTS backend to use.") parser.add_argument("--tts-backend", default=default_arguments['tts-backend'], help="Specifies which TTS backend to use.")
@ -2137,6 +2139,7 @@ def get_default_settings( hypenated=True ):
'output-sample-rate': args.output_sample_rate, 'output-sample-rate': args.output_sample_rate,
'autocalculate-voice-chunk-duration-size': args.autocalculate_voice_chunk_duration_size, 'autocalculate-voice-chunk-duration-size': args.autocalculate_voice_chunk_duration_size,
'output-volume': args.output_volume, 'output-volume': args.output_volume,
'results-folder': args.results_folder,
'tts-backend': args.tts_backend, 'tts-backend': args.tts_backend,
@ -2183,6 +2186,7 @@ def update_args( **kwargs ):
args.output_sample_rate = 44000 args.output_sample_rate = 44000
args.autocalculate_voice_chunk_duration_size = settings['autocalculate_voice_chunk_duration_size'] args.autocalculate_voice_chunk_duration_size = settings['autocalculate_voice_chunk_duration_size']
args.output_volume = settings['output_volume'] args.output_volume = settings['output_volume']
args.results_folder = settings['results_folder']
args.tts_backend = settings['tts_backend'] args.tts_backend = settings['tts_backend']

12
src/webui.py
View File

@ -120,7 +120,7 @@ def update_training_configs():
def history_view_results( voice ): def history_view_results( voice ):
results = [] results = []
files = [] files = []
outdir = f"./results/{voice}/" outdir = f"./{args.results_folder}/{voice}/"
for i, file in enumerate(sorted(os.listdir(outdir))): for i, file in enumerate(sorted(os.listdir(outdir))):
if file[-4:] != ".wav": if file[-4:] != ".wav":
continue continue
@ -281,11 +281,11 @@ def update_voices():
return ( return (
gr.Dropdown.update(choices=get_voice_list(append_defaults=True)), gr.Dropdown.update(choices=get_voice_list(append_defaults=True)),
gr.Dropdown.update(choices=get_voice_list()), gr.Dropdown.update(choices=get_voice_list()),
gr.Dropdown.update(choices=get_voice_list("./results/")), gr.Dropdown.update(choices=get_voice_list(args.results_folder)),
) )
def history_copy_settings( voice, file ): def history_copy_settings( voice, file ):
return import_generate_settings( f"./results/{voice}/{file}" ) return import_generate_settings( f"./{args.results_folder}/{voice}/{file}" )
def setup_gradio(): def setup_gradio():
global args global args
@ -309,7 +309,7 @@ def setup_gradio():
voice_list_with_defaults = get_voice_list(append_defaults=True) voice_list_with_defaults = get_voice_list(append_defaults=True)
voice_list = get_voice_list() voice_list = get_voice_list()
result_voices = get_voice_list("./results/") result_voices = get_voice_list(args.results_folder)
autoregressive_models = get_autoregressive_models() autoregressive_models = get_autoregressive_models()
diffusion_models = get_diffusion_models() diffusion_models = get_diffusion_models()
@ -571,6 +571,8 @@ def setup_gradio():
EXEC_SETTINGS['autocalculate_voice_chunk_duration_size'] = gr.Number(label="Auto-Calculate Voice Chunk Duration (in seconds)", precision=0, value=args.autocalculate_voice_chunk_duration_size) EXEC_SETTINGS['autocalculate_voice_chunk_duration_size'] = gr.Number(label="Auto-Calculate Voice Chunk Duration (in seconds)", precision=0, value=args.autocalculate_voice_chunk_duration_size)
EXEC_SETTINGS['output_volume'] = gr.Slider(label="Output Volume", minimum=0, maximum=2, value=args.output_volume) EXEC_SETTINGS['output_volume'] = gr.Slider(label="Output Volume", minimum=0, maximum=2, value=args.output_volume)
EXEC_SETTINGS['device_override'] = gr.Textbox(label="Device Override", value=args.device_override) EXEC_SETTINGS['device_override'] = gr.Textbox(label="Device Override", value=args.device_override)
EXEC_SETTINGS['results_folder'] = gr.Textbox(label="Results Folder", value=args.results_folder)
with gr.Column(): with gr.Column():
# EXEC_SETTINGS['tts_backend'] = gr.Dropdown(TTSES, label="TTS Backend", value=args.tts_backend if args.tts_backend else TTSES[0]) # EXEC_SETTINGS['tts_backend'] = gr.Dropdown(TTSES, label="TTS Backend", value=args.tts_backend if args.tts_backend else TTSES[0])
@ -651,7 +653,7 @@ def setup_gradio():
] ]
) )
history_results_list.change( history_results_list.change(
fn=lambda voice, file: f"./results/{voice}/{file}", fn=lambda voice, file: f"./{args.results_folder}/{voice}/{file}",
inputs=[ inputs=[
history_voices, history_voices,
history_results_list, history_results_list,