Return the proper normalization module. If causal is True, this will ensure the returned module is causal, or return an error if the normalization

Push the given symbol on the stream, flushing out bits if possible. Args: symbol (int): symbol to encode with the AC.

Turn the given PDF into a quantized CDF that splits [0, 2 ** self.total_range_bits - 1] into chunks of size roughly proportional to the PDF.

See `pad_for_conv1d`.

Tiny wrapper around F.pad, just to allow for reflect padding on small input. If this is the case, we insert extra 0 padding to the right before th

Pull a symbol, reading as many bits from the stream as required. This returns `None` when the stream has been exhausted. Args:

Remove padding from x, handling properly zero padding. Only for 1d!

Returns the number of steps trained from a checkpoint based on the filename. Filename format assumed to be something like "/path/to/soundstorm.20

Parameters ---------- codes : torch.tensor Indices for each quantizer. Shape: (n_q, batch, timesteps). s

Parameters ---------- x : torch.tensor Input wavs. Shape: (batch, channels, timesteps). n_q : int, optio

Flush the remaining information to the stream.

Parameters ---------- config_path : str Path of model configuration file. ckpt_path : str Pa

Parameters ---------- config : json Model Config.

Average a dictionary of metrics across all workers, using the optional `count` as unormalized weight.

Broadcast the tensors from the given parameters to all workers. This can be used to ensure that all workers have the same model to start with.

Decode the given codes to the quantized representation. Args: codes (torch.Tensor): Input indices for each quantizer.