TypeError: repeat(): argument 'repeats' must be tuple of ints, but found element of type float at pos 2

[e0xextazy]

To reproduce: run codes/models/audio/tts/unified_voice3.py.

In:

ac_expansion_factor = mel_codes.shape[-1] / aligned_codes.shape[-1]
aligned_codes = aligned_codes.repeat(1, ac_expansion_factor)

ac_expansion_factor = 0.4332 = 250 / 577 but should be Int

[e0xextazy]

i want to test solo model by random inputs

[neonbjb]

unified_voice3 is an experimental model that will probably be deleted. Assuming you are looking to train tortoise, you want unified_voice2.

[e0xextazy]

Thank you!