azalea/VITS-fast-fine-tuning
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

upload files

Browse Source
This commit is contained in:
Plachta
2023-02-14 18:12:41 +08:00
parent 0d46fbd5c2
commit eb546d3607
1 changed files with 8 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
VC_inference.py
+8 -8
View File
@@ -14,7 +14,7 @@ from models_infer import SynthesizerTrn
from text import text_to_sequence from text import text_to_sequence
import gradio as gr import gradio as gr
import torchaudio import torchaudio
device = "cuda:0" if torch.cuda.is_available() else "cpu"
def get_text(text, hps): def get_text(text, hps):
text_norm = text_to_sequence(text, hps.symbols, hps.data.text_cleaners) text_norm = text_to_sequence(text, hps.symbols, hps.data.text_cleaners)
if hps.data.add_blank: if hps.data.add_blank:
@@ -37,14 +37,14 @@ def create_vc_fn(model, hps, speaker_ids):
if sampling_rate != hps.data.sampling_rate: if sampling_rate != hps.data.sampling_rate:
audio = librosa.resample(audio, orig_sr=sampling_rate, target_sr=hps.data.sampling_rate) audio = librosa.resample(audio, orig_sr=sampling_rate, target_sr=hps.data.sampling_rate)
with no_grad(): with no_grad():
y = torch.FloatTensor(audio) y = torch.FloatTensor(audio).to(device)
y = y.unsqueeze(0) y = y.unsqueeze(0)
spec = spectrogram_torch(y, hps.data.filter_length, spec = spectrogram_torch(y, hps.data.filter_length,
hps.data.sampling_rate, hps.data.hop_length, hps.data.win_length, hps.data.sampling_rate, hps.data.hop_length, hps.data.win_length,
center=False) center=False).to(device)
spec_lengths = LongTensor([spec.size(-1)]) spec_lengths = LongTensor([spec.size(-1)]).to(device)
sid_src = LongTensor([original_speaker_id]) sid_src = LongTensor([original_speaker_id]).to(device)
sid_tgt = LongTensor([target_speaker_id]) sid_tgt = LongTensor([target_speaker_id]).to(device)
audio = model.voice_conversion(spec, spec_lengths, sid_src=sid_src, sid_tgt=sid_tgt)[0][ audio = model.voice_conversion(spec, spec_lengths, sid_src=sid_src, sid_tgt=sid_tgt)[0][
0, 0].data.cpu().float().numpy() 0, 0].data.cpu().float().numpy()
del y, spec, spec_lengths, sid_src, sid_tgt del y, spec, spec_lengths, sid_src, sid_tgt
@@ -54,11 +54,11 @@ def create_vc_fn(model, hps, speaker_ids):
if __name__ == "__main__": if __name__ == "__main__":
parser = argparse.ArgumentParser() parser = argparse.ArgumentParser()
parser.add_argument("--model_dir", default="./G_latest.pth", help="directory to your fine-tuned model") parser.add_argument("--model_dir", default="./G_latest.pth", help="directory to your fine-tuned model")
parser.add_argument("--share", default=True, help="make link public (used in colab)") parser.add_argument("--share", default=False, help="make link public (used in colab)")
args = parser.parse_args() args = parser.parse_args()
hps = utils.get_hparams_from_file("./configs/finetune_speaker.json") hps = utils.get_hparams_from_file("./configs/finetune_speaker.json")
device = "cpu"
net_g = SynthesizerTrn( net_g = SynthesizerTrn(
len(hps.symbols), len(hps.symbols),