This repository has been archived by the owner on Oct 31, 2023. It is now read-only.

Denoise an audio array (ndarray) instead of an audio file (mp3) #161

Open

d4rkc0de opened this issue Oct 7, 2023 · 0 comments

d4rkc0de commented Oct 7, 2023

I have an audio array of type ndarray that I want to denoise, in the example notebooks it uses a mp3 file:

wav, sr = torchaudio.load('alex_noisy.mp3')
Can I do the same with an audio array ?

I tried:

audio_array = generate_audio(text_prompt)
wav = torch.from_numpy(audio_array)
wav = convert_audio(wav.cuda(), SAMPLE_RATE, model.sample_rate, model.chin)

but I got this error:
AttributeError: 'Tensor' object has no attribute 'channels'

The text was updated successfully, but these errors were encountered:

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.