Conv1d with some builtin handling of asymmetric or causal padding and normalization.

Allow simple CUDA Graphing of a function. Args: func: callable, taking any number of arguments. Its tensors arguments should

Residual block from SEANet model. Args: dim (int): Dimension of the input/output. kernel_sizes (list): List of kernel sizes for t

ConvTranspose1d with some builtin handling of asymmetric or causal padding and normalization.

Input for waveform based conditionings. Wav should always be 3-dim `[B, C, T]` even before collation.

Standard class for representing the set of potential inputs to the conditioners. Typically, `audiocraft.data.audio_dataset.SegmentInfo` will conve

Return type for a conditioner: both a condition tensor, and a mask indicating valid positions.

Transformer with Streaming / Causal support. Args: d_model (int): Dimension of the data. num_heads (int): Number of heads.

Layer scale from [Touvron et al 2021] (https://arxiv.org/pdf/2103.17239.pdf). This rescales diagonally the residual outputs close to 0, with a lea

Residual Vector Quantizer. Args: dimension (int): Dimension of the codebooks. input_dimension (None or int): dimension of the inp

SEANet decoder. Args: channels (int): Audio channels. dimension (int): Intermediate representation dimension. n_filters (

Gating FFN layer, using the given activation. Args: dim (int): dimension of the input and output of the transformer. activati

Contains the paths to each sub model, along with some extra configuration. Args: moshi_weights: path to the checkpoint for the Moshi

Condition fuser handles the logic to combine the different conditions to the actual model input. Args: fuse2cond (tp.Dict[str, str]):

Prepare and provide conditions given all the supported conditioners. Args: conditioners (dict): Dictionary of conditioners. devic

Downsampling by some integer amount `stride` using convolutions with a kernel size of twice the stride. If `causal` is True, the output u

Upsample by some integer amount `stride` using transposed convolutions.

Codebook with Euclidean distance. Args: dim (int): Dimension. codebook_size (int): Codebook size. decay (float): Decay fo

Transformer-based language model on multiple streams of codes. Args: n_q (int): Number of parallel streams to model as input. dep

Lookup table TextConditioner. Args: n_bins (int): Number of bins. dim (int): Hidden dim of the model (text-encoder/LUT).

Mimi model operating on the raw waveform. Args: encoder (nn.Module): Encoder network. decoder (nn.Module): Decoder network.

This tokenizer should be used for global conditioners such as: artist, genre, key, etc. The difference between this and WhiteSpaceTokenizer is tha

Wrapper around Conv1d and normalization applied to this conv to provide a uniform interface across normalization approaches.

Wrapper around ConvTranspose1d and normalization applied to this conv to provide a uniform interface across normalization approaches.

Residual vector quantization implementation. Follows Algorithm 1. in https://arxiv.org/pdf/2107.03312.pdf

Efficient streaming KVCache to be compatible with Cuda Graph. Args: batch_size (int): Batch size. num_heads (int): Number of head

Rotary positional embedding (RoPE) from [Su et al 2022](https://arxiv.org/abs/2104.09864). Args: max_period (float): Maximum period of th

SEANet encoder. Args: channels (int): Audio channels. dimension (int): Intermediate representation dimension. n_filters (

Residual Vector Quantizer with separate projections for the first quantizer and the rest. Args: n_q (int): Number of residual vector quan

Base State for streaming, requires to be resetable and also support the context protocol. The state will be entered when

Similar to `nn.MultiheadAttention` but with support for streaming, causal evaluation. Args: embed_dim (int): Dimension to project to.

Vector quantization implementation. Currently supports only euclidean distance. Args: dim (int): Dimension codebook_size (int

Dropout with a given probability per attribute. This is different from the behavior of ClassifierFreeGuidanceDropout as this allows for attributes

Base model for all conditioner modules. Args: dim (int): internal dim of the model. output_dim (int): Output dim of the condition

Base class for quantizers.

Classifier Free Guidance dropout. All attributes are dropped with the same probability. Args: p (float): Probability to apply conditi

Base API for all compression model that aim at being used as audio tokenizers with a language model.

Base module for all dropout modules.

Fake quantizer that actually does not perform any quantization.