How to convert audio (e.g. wav) to tensor and back?

audio
1

Hello!
It is easy to convert audio (wav) to tensor using .load() command. But I am interested in reverse operation. Can someone help me?

Thanks!

2

Isn’t it torchaudio.save('amazing_sound.wav', sound, sample_rate) ?

By the way, you can find the doc here: http://pytorch.org/audio/

3

I assume there isn’t any way to convert the tensor back to byte format without saving. That’s a feature worth adding I guess.