m

m

m

torchaudio.save("t.wav", tensor.unsqueeze(0), sampling_rate, bits_per_sample=16)

m

m

m

m

m

48000 hz

update audio

m

add train.py

m

m

ｍ

speaker

m

m

init

initial commit