[BUG FIX] [Stable Audio Pipeline] Resolve torch.Tensor.new_zeros() TypeError in function prepare_latents caused by audio_vae_length

[syntaxticsugr]

What does this PR do?

Parameter initial_audio_waveforms when passed torch.Tensor as described here raises TypeError: new_zeros(): argument 'size' failed to unpack the object at pos 3 with error "type must be tuple of ints,but got float"

Notebook to Reproduce the Error

In function prepare_latents:

audio_vae_length = self.transformer.config.sample_size * self.vae.hop_length

audio_vae_length evaluates to a float because self.transformer.config.sample_size returns a float

audio = initial_audio_waveforms.new_zeros(audio_shape)

torch.Tensor.new_zeros() accepts a single argument size – a list, tuple, or torch.Size of integers defining the shape of the output tensor. Source

audio_shape = (batch_size // num_waveforms_per_prompt, audio_channels, audio_vae_length)

But audio_shape is of type – (int, int, float) because audio_vae_length is a float

Proposed Fix is to wrap self.transformer.config.sample_size with int()

Notebook with the Applied Fix

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline?
  Do not work on the main branch. 🤦‍♂️
• Did you read our philosophy doc (important for complex PRs)?
• Was this discussed/approved via a GitHub issue or the forum? Please add a link to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
• Did you write any new necessary tests?

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

[hlky]

Thanks @syntaxticsugr!