Medix-AI/coqui-tts
Medix-AI
/
coqui-tts
mirror of https://github.com/coqui-ai/TTS.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

fix(bin.extract_tts_spectrograms): set quantization bits

This commit is contained in:
Enno Hermann 2023-11-14 14:07:36 +01:00
parent 13e640f17e
commit aa0fbdf27e
1 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
TTS/bin/extract_tts_spectrograms.py
View File

@ -159,7 +159,7 @@ def inference(
def extract_spectrograms( def extract_spectrograms(
data_loader, model, ap, output_path, quantized_wav=False, save_audio=False, debug=False, metada_name="metada.txt" data_loader, model, ap, output_path, quantize_bits=0, save_audio=False, debug=False, metada_name="metada.txt"
): ):
model.eval() model.eval()
export_metadata = [] export_metadata = []
@ -196,8 +196,8 @@ def extract_spectrograms(
_, wavq_path, mel_path, wav_gl_path, wav_path = set_filename(wav_file_path, output_path) _, wavq_path, mel_path, wav_gl_path, wav_path = set_filename(wav_file_path, output_path)
# quantize and save wav # quantize and save wav
if quantized_wav: if quantize_bits > 0:
wavq = ap.quantize(wav) wavq = ap.quantize(wav, quantize_bits)
np.save(wavq_path, wavq) np.save(wavq_path, wavq)
# save TTS mel # save TTS mel
@ -263,7 +263,7 @@ def main(args): # pylint: disable=redefined-outer-name
model, model,
ap, ap,
args.output_path, args.output_path,
quantized_wav=args.quantized, quantize_bits=args.quantize_bits,
save_audio=args.save_audio, save_audio=args.save_audio,
debug=args.debug, debug=args.debug,
metada_name="metada.txt", metada_name="metada.txt",
@ -277,7 +277,7 @@ if __name__ == "__main__":
parser.add_argument("--output_path", type=str, help="Path to save mel specs", required=True) parser.add_argument("--output_path", type=str, help="Path to save mel specs", required=True)
parser.add_argument("--debug", default=False, action="store_true", help="Save audio files for debug") parser.add_argument("--debug", default=False, action="store_true", help="Save audio files for debug")
parser.add_argument("--save_audio", default=False, action="store_true", help="Save audio files") parser.add_argument("--save_audio", default=False, action="store_true", help="Save audio files")
parser.add_argument("--quantized", action="store_true", help="Save quantized audio files") parser.add_argument("--quantize_bits", type=int, default=0, help="Save quantized audio files if non-zero")
parser.add_argument("--eval", type=bool, help="compute eval.", default=True) parser.add_argument("--eval", type=bool, help="compute eval.", default=True)
args = parser.parse_args() args = parser.parse_args()