Decode the given codes to a reconstructed representation. Args: codes (torch.Tensor): Int tensor of shape [B, K, T] Retu

Encode the given input tensor to quantized representation. Args: x (torch.Tensor): Float tensor of shape [B, C, T] Retur

Reset the streaming state.

Create normalization module for transformer encoder layer. Args: norm_type (str): Normalization method. dim (int): Dimension of t

Given a tensor of codes of shape `[*]`, returns a tensor of shape `[*, D]`, corresponding to the centroids associated to each code index.

torch.multinomial with arbitrary number of dimensions, and number of candidates on the last dimension. Args: input (torch.Tensor): The in

Returns a 0. value on the given device without introducing a synchronization point.

Decode from the discrete codes to continuous latent space.

Check if obj is an instance of a dataclass (not the class itself). Parameters ---------- obj : Any Object to check.

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

Context manager to enter streaming mode. Reset streaming state on exit.

Projects a batch of waveforms to unquantized latent space. Args: x (torch.Tensor): Float tensor of shape [B, C, T]. Retu

Renormalizes values from `rvq_first` and `rvq_rest` and adds them. This allows correcting statistics that are normalized by the number of qua

Restores the streaming state from the given `state_dict` dict looking up fields by adding `prefix` to each key in `keys` to look up values.

Restore the streaming state from the given pt_state dict Parameters ---------- value : torch.Tensor Specific streaming state

Run command, return True on success.

Set the streaming state in-place from the given `state_dict` dict. Parameters ---------- streaming_state : State Specific streami

Shared core for stepping through the voice prompt. This generator yields at each *checkpoint* where the async wrapper may want to con

Decode the given codes to the quantized representation.

Encode a given input tensor with the specified frame rate at the given bandwidth. The RVQ encode method sets the appropriate number of quantiz

Given a tensor `x` of shape `[*, D]`, returns a tensor of integer codes of shape `[*]`. The codes are defined as the indexes of the centroids

See `pad_for_conv1d`.

get_lan_ip() Get the LAN IP address of the current machine. Returns ------- str LAN IP address as a string.

Utility to apply a multi linear layer to the given input. A multi linear layer applies a different set of weight for each time step. Args:

Disable torch.compile locally. Now Pytorch 2.4 provides a function to do that.

Given logits of shape [*, Card], returns a LongTensor of shape [*].