# Copyright 2024 Meta Platforms, Inc. and affiliates, and the HuggingFace Inc. team. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. """Mimi model configuration""" import math import numpy as np from huggingface_hub.dataclasses import strict from ...configuration_utils import PreTrainedConfig from ...modeling_rope_utils import RopeParameters from ...utils import auto_docstring @auto_docstring(checkpoint="kyutai/mimi") @strict class MimiConfig(PreTrainedConfig): r""" audio_channels (`int`, *optional*, defaults to 1): Number of channels in the audio data. Either 1 for mono or 2 for stereo. num_filters (`int`, *optional*, defaults to 64): Number of convolution kernels of first `MimiConv1d` down sampling layer. num_residual_layers (`int`, *optional*, defaults to 1): Number of residual layers. upsampling_ratios (`Sequence[int]`, *optional*): Kernel size and stride ratios. The encoder uses downsampling ratios instead of upsampling ratios, hence it will use the ratios in the reverse order to the ones specified here that must match the decoder order. If not specified, will defaults to `[8, 6, 5, 4]` last_kernel_size (`int`, *optional*, defaults to 3): Kernel size for the last convolution layer. residual_kernel_size (`int`, *optional*, defaults to 3): Kernel size for the residual layers. dilation_growth_rate (`int`, *optional*, defaults to 2): How much to increase the dilation with each layer. use_causal_conv (`bool`, *optional*, defaults to `True`): Whether to use fully causal convolution. pad_mode (`str`, *optional*, defaults to `"constant"`): Padding mode for the convolutions. compress (`int`, *optional*, defaults to 2): Reduced dimensionality in residual branches. trim_right_ratio (`float`, *optional*, defaults to 1.0): Ratio for trimming at the right of the transposed convolution under the `use_causal_conv = True` setup. If equal to 1.0, it means that all the trimming is done at the right. num_quantizers (`int`, *optional*, defaults to 32): Number of quantizer channels, or codebooks, in the quantizer. use_conv_shortcut (`bool`, *optional*, defaults to `False`): Whether to use a convolutional layer as the 'skip' connection in the `MimiResnetBlock` block. If False, an identity function will be used, giving a generic residual connection. vector_quantization_hidden_dimension (`int`, *optional*, defaults to 256): Intermediate representation dimension in the residual vector quantization space. num_semantic_quantizers (`int`, *optional*, defaults to 1): Number of semantic quantizer channels, or codebooks, in the semantic quantizer. Must be lower than `num_quantizers`. upsample_groups (`int`, *optional*, defaults to 512): If `frame_rate!=encodec_frame_rate`, indicates the number of groups used in the upsampling operation to go from one rate to another. use_streaming (`bool`, *optional*, defaults to `False`): Whether to use streaming mode. If `True`, the model encode method will return the padding cache that can be used in a subsequent call to the encode method. Example: ```python >>> from transformers import MimiModel, MimiConfig >>> # Initializing a "kyutai/mimi" style configuration >>> configuration = MimiConfig() >>> # Initializing a model (with random weights) from the "kyutai/mimi" style configuration >>> model = MimiModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```""" model_type = "mimi" sampling_rate: int = 24_000 audio_channels: int = 1 hidden_size: int = 512 num_filters: int = 64 num_residual_layers: int = 1 upsampling_ratios: list[int] | None = None kernel_size: int = 7 last_kernel_size: int = 3 residual_kernel_size: int = 3 dilation_growth_rate: int = 2 use_causal_conv: bool = True pad_mode: str = "constant" compress: int = 2 trim_right_ratio: float = 1.0 codebook_size: int = 2048 codebook_dim: int = 256 num_quantizers: int = 32 use_conv_shortcut: bool = False vector_quantization_hidden_dimension: int = 256 num_semantic_quantizers: int = 1 upsample_groups: int = 512 num_hidden_layers: int = 8 intermediate_size: int = 2048 num_attention_heads: int = 8 num_key_value_heads: int = 8 head_dim: int | None = None hidden_act: str = "gelu" max_position_embeddings: int = 8000 initializer_range: float = 0.02 norm_eps: float = 1e-5 use_cache: bool = False use_streaming: bool = False rope_parameters: RopeParameters | dict | None = None sliding_window: int = 250 attention_dropout: float | int = 0.0 layer_scale_initial_scale: float = 0.01 attention_bias: bool = False tie_word_embeddings: bool = True def __post_init__(self, **kwargs): self.upsampling_ratios = self.upsampling_ratios if self.upsampling_ratios else [8, 6, 5, 4] self.codebook_dim = self.codebook_dim if self.codebook_dim is not None else self.hidden_size self.head_dim = self.head_dim or self.hidden_size // self.num_attention_heads # Handle backward compatibility for frame_rate: # If frame_rate is explicitly provided, use it (backward compatibility) # Otherwise, compute it from other parameters (correctly) self._frame_rate = kwargs.pop("frame_rate", None) super().__post_init__(**kwargs) def validate_architecture(self): """Part of `@strict`-powered validation. Validates the architecture of the config.""" if self.num_semantic_quantizers >= self.num_quantizers: raise ValueError( f"The number of semantic quantizers should be lower than the total number of quantizers {self.num_quantizers}, but is currently {self.num_semantic_quantizers}." ) @property def encodec_frame_rate(self) -> int: hop_length = np.prod(self.upsampling_ratios) return math.ceil(self.sampling_rate / hop_length) @property def num_codebooks(self) -> int: # alias to num_quantizers return self.num_quantizers @property def frame_size(self) -> int: # 1. we need each encoder conv stride # first conv strides = [1] # layer convs for ratio in reversed(self.upsampling_ratios): for j in range(self.num_residual_layers): len_kernel_sizes = len(self.residual_kernel_size) if isinstance(self.residual_kernel_size, list) else 1 strides.extend([1] * (len_kernel_sizes + 1)) if self.use_conv_shortcut: # skip connection strides.append(1) strides.append(ratio) # last conv strides.append(1) # downsampling layer strides.append(2) return math.prod(strides) @property def frame_rate(self) -> float: # handle backward compatibility if self._frame_rate is not None: return self._frame_rate return self.sampling_rate / self.frame_size __all__ = ["MimiConfig"]