Where is the output saved?

[4DeepThought2]

I feel like an idiot here. Running this on Windows (WSL with conda). Where in the heck does the audio file output get saved?

[nolanblew]

The output doesn't seem to be saved anywhere - just in memory.

You can save the output by first importing soundfile:

import soundfile as sf

Then at the end of the script, put:

sf.write('./output.wav', output_audio, sr) # replace output.wav with your output path if you want to change it

If you had a lot of trouble like I did getting very small or empty audio, make sure that you have your transcript filled out correctly, or set deep_clone to False (though in my experiments, the results were not anywhere close to the reference audio)

[akshhack]

Thanks for the answer @nolanblew !