The Moshi model was proposed in Moshi: a speech-text foundation model for real-time dialogue by Alexandre Défossez, Laurent Mazaré, Manu Orsini, Amélie Royer, Patrick Pérez, Hervé Jégou, Edouard Grave and Neil Zeghidour.
Moshi is a speech-text foundation model that casts spoken dialogue as speech-to-speech generation. Starting from a text language model backbone, Moshi generates speech as tokens from the residual quantizer of a neural audio codec, while modeling separately its own speech and that of the user into parallel streams. This allows for the removal of explicit speaker turns, and the modeling of arbitrary conversational dynamics. Moshi also predicts time-aligned text tokens as a prefix to audio tokens. This “Inner Monologue” method significantly improves the linguistic quality of generated speech and provides streaming speech recognition and text-to-speech. As a result, Moshi is the first real-time full-duplex spoken large language model, with a theoretical latency of 160ms, 200ms in practice.
The abstract from the paper is the following:
We introduce Moshi, a speech-text foundation model and full-duplex spoken dialogue framework. Current systems for spoken dialogue rely on pipelines of independent components, namely voice activity detection, speech recognition, textual dialogue and text-to-speech. Such frameworks cannot emulate the experience of real conversations. First, their complexity induces a latency of several seconds between interactions. Second, text being the intermediate modality for dialogue, non-linguistic information that modifies meaning— such as emotion or non-speech sounds— is lost in the interaction. Finally, they rely on a segmentation into speaker turns, which does not take into account overlapping speech, interruptions and interjections. Moshi solves these independent issues altogether by casting spoken dialogue as speech-to-speech generation. Starting from a text language model backbone, Moshi generates speech as tokens from the residual quantizer of a neural audio codec, while modeling separately its own speech and that of the user into parallel streams. This allows for the removal of explicit speaker turns, and the modeling of arbitrary conversational dynamics. We moreover extend the hierarchical semantic-to-acoustic token generation of previous work to first predict time-aligned text tokens as a prefix to audio tokens. Not only this “Inner Monologue” method significantly improves the linguistic quality of generated speech, but we also illustrate how it can provide streaming speech recognition and text-to-speech. Our resulting model is the first real-time full-duplex spoken large language model, with a theoretical latency of 160ms, 200ms in practice, and is available at github.com/kyutai-labs/moshi.
Moshi deals with 3 streams of information:
Similarly to ~MusicgenModel, audio is represented with audio codebooks, which can be interpreted like tokens. The main difference between text tokens and audio codebooks is that audio codebooks introduce an additional dimension of information. Text tokens are typically of dim (batch_size, sequence_length) but audio tokens are of dim (batch_size, num_codebooks, sequence_length).
Moshi’s made of 3 components:
1. The main decoder (Helium in the paper)
It corresponds to MoshiForCausalLM. It is strictly a classic text LLM, that uses an architecture similar to ~GemmaForCausalLM. In other words, it takes text tokens, embeds them, pass them through the decoder and a language head, to get text logits.
2. The depth decoder
On its own, it’s also a classic LLM, but this time, instead of generating over the time dimension, it generates over the codebook dimension.
It also means that its context length is num_codebooks, thus it can’t generate more than num_codebooks.
Note that each timestamp - i.e each codebook - gets its own set of Linear Layers and Embeddings.
3. MimiModel
It’s the audio encoder from Kyutai, that has recently been integrated to transformers, which is used to “tokenize” audio. It has the same use that ~EncodecModel has in ~MusicgenModel.
Tips:The original checkpoints can be converted using the conversion script src/transformers/models/moshi/convert_moshi_transformers.py
This implementation has two main aims:
It is designed for intermediate use. We strongly recommend using the original implementation to infer the model in real-time streaming.
1. Model generation
Moshi is a streaming auto-regressive model with two streams of audio. To put it differently, one audio stream corresponds to what the model said/will say and the other audio stream corresponds to what the user said/will say.
MoshiForConditionalGeneration.generate() thus needs 3 inputs:
input_ids - corresponding to the text token historymoshi_input_values or moshi_audio_codes- corresponding to the model audio historyuser_input_values or user_audio_codes - corresponding to the user audio historyThese three inputs must be synchronized. Meaning that their lengths must correspond to the same number of tokens.
You can dynamically use the 3 inputs depending on what you want to test:
input_ids can be filled with pad tokens and user_input_values can be a zero tensor of the same shape than the user prompt.The original model is synchronized text with audio by padding the text in between each token enunciation.
To follow the example of the following image, "Hello, I'm Moshi" could be transformed to "Hello,<pad><unk>I'm Moshi".
MoshiForConditionalGeneration.generate() then auto-regressively feeds to itself its own audio stream, but since it doesn’t have access to the user input stream while using transformers, it will thus assume that the user is producing blank audio.
>>> from datasets import load_dataset, Audio
>>> import torch, math
>>> from transformers import MoshiForConditionalGeneration, AutoFeatureExtractor, AutoTokenizer
>>> librispeech_dummy = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
>>> feature_extractor = AutoFeatureExtractor.from_pretrained("kyutai/moshiko-pytorch-bf16")
>>> tokenizer = AutoTokenizer.from_pretrained("kyutai/moshiko-pytorch-bf16")
>>> device = "cuda"
>>> dtype = torch.bfloat16
>>>
>>> librispeech_dummy = librispeech_dummy.cast_column("audio", Audio(sampling_rate=feature_extractor.sampling_rate))
>>> audio_sample = librispeech_dummy[-1]["audio"]["array"]
>>> user_input_values = feature_extractor(raw_audio=audio_sample, sampling_rate=feature_extractor.sampling_rate, return_tensors="pt").to(device=device, dtype=dtype)
>>>
>>> moshi_input_values = torch.zeros_like(user_input_values.input_values)
>>>
>>> num_tokens = math.ceil(moshi_input_values.shape[-1] * waveform_to_token_ratio)
>>> input_ids = torch.ones((1, num_tokens), device=device, dtype=torch.int64) * tokenizer.encode("<pad>")[0]
>>>
>>> output = model.generate(input_ids=input_ids, user_input_values=user_input_values.input_values, moshi_input_values=moshi_input_values, max_new_tokens=25)
>>> text_tokens = output.sequences
>>> audio_waveforms = output.audio_sequences
2. Model training
Most of the work has to be done during data creation/pre-processing, because of the need to align/synchronize streams.
Once it’s done, you can simply forward text_labels and audio_labels to MoshiForConditionalGeneration.forward(), alongside the usual inputs, to get the model loss.
A training guide will come soon, but user contributions are welcomed!
How does the model forward the inputs / generate:The input streams are embedded and combined into inputs_embeds.
inputs_embeds is passed through the main decoder, which processes it like a normal LLM would.
The main decoder outputs text logits but also its last hidden state which is called temporal context in the paper.
The depth decoder switches the dimension on which we forward / generate (codebooks instead of time). It uses the token generated from text logits and the temporal context to auto-regressively generate audio codebooks.
This model was contributed by Yoach Lacombe (ylacombe).
The original code can be found here.
MoshiConfig class transformers.MoshiConfig < source >( vocab_size = 32000 hidden_size = 4096 num_hidden_layers = 32 num_attention_heads = 32 num_key_value_heads = None audio_vocab_size = None max_position_embeddings = 3000 rope_theta = 10000.0 hidden_act = 'silu' head_dim = None initializer_range = 0.02 use_cache = True sliding_window = 3000 attention_dropout = 0.0 ffn_dim = 22528 rms_norm_eps = 1e-08 num_codebooks = 8 tie_word_embeddings = False **kwargs )
Parameters
int, optional, defaults to 32000) — Vocabulary size of the MoshiDecoder model. Defines the number of different tokens that can be represented by the inputs_ids passed when calling MoshiDecoder. int, optional, defaults to 4096) — Dimensionality of the layers and the pooler layer of the main decoder. int, optional, defaults to 32) — Number of decoder layers. int, optional, defaults to 32) — Number of attention heads for each attention layer in the main decoder block. int, optional) — This is the number of key_value heads that should be used to implement Grouped Query Attention. If num_key_value_heads=num_attention_heads, the model will use Multi Head Attention (MHA), if num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed by meanpooling all the original heads within that group. For more details checkout this paper. If it is not specified, will default to num_attention_heads. int, optional) — Vocabulary size of the audio part of model. Defines the number of different tokens that can be represented by the audio_codes passed when calling the Moshi models. int, optional, defaults to 3000) — The maximum sequence length that this model might ever be used with. Typically, set this to something large just in case (e.g., 512 or 1024 or 2048). float, optional, defaults to 10000.0) — The base period of the RoPE embeddings. str or function, optional, defaults to "silu") — The non-linear activation function (function or string) in the decoder. int, optional, defaults to hidden_size // num_attention_heads) — The attention head dimension. float, optional, defaults to 0.02) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices. bool, optional, defaults to True) — Whether or not the model should return the last key/values attentions (not used by all models). Only relevant if config.is_decoder=True. int, optional, defaults to 3000) — Sliding window attention window size. If not specified, will default to 3000. float, optional, defaults to 0.0) — The dropout ratio for the attention probabilities. int, optional, defaults to 22528) — Dimensionality of the “intermediate” (often named feed-forward) layer in the main decoder block. Must be even. float, optional, defaults to 1e-08) — The epsilon used by the rms normalization layers. int, optional, defaults to 8) — The number of audio codebooks for each audio channels. bool, optional, defaults to False) — Whether to tie weight embeddings This is the configuration class to store the configuration of a MoshiModel. It is used to instantiate a Moshi model according to the specified arguments, defining the audio encoder, Moshi depth decoder and Moshi decoder configs. Instantiating a configuration with the defaults will yield a similar configuration to that of the Moshiko model, e.g. kmhf/hf-moshiko
Configuration objects inherit from PretrainedConfig and can be used to control the model outputs. Read the documentation from PretrainedConfig for more information.
Example:
>>> from transformers import (
... MoshiConfig,
... MoshiForConditionalGeneration,
... )
>>> configuration = MoshiConfig()
>>>
>>> model = MoshiForConditionalGeneration(configuration)
>>>
>>> configuration = model.config
>>>
>>> model.save_pretrained("kmhf/hf-moshiko")
>>>
>>> moshi_config = MoshiConfig.from_pretrained("kmhf/hf-moshiko")
>>> model = MoshiForConditionalGeneration.from_pretrained("kmhf/hf-moshiko", config=moshi_config) from_audio_encoder_config < source >
( audio_encoder_config: PretrainedConfig **kwargs ) → MoshiConfig
An instance of a configuration object
Instantiate a MoshiConfig (or a derived class) from an audio encoder configuration.
MoshiDepthConfig class transformers.MoshiDepthConfig < source >( vocab_size = 32000 hidden_size = 1024 input_size = 4096 num_hidden_layers = 6 num_attention_heads = 16 num_key_value_heads = None audio_vocab_size = 2048 max_position_embeddings = 9 hidden_act = 'silu' head_dim = None initializer_range = 0.02 use_cache = True sliding_window = 8 attention_dropout = 0.0 ffn_dim = 5632 rms_norm_eps = 1e-08 num_codebooks = 8 tie_word_embeddings = False **kwargs )
Parameters
int, optional, defaults to 32000) — Vocabulary size of the MoshiDepthDecoder model. Defines the number of different tokens that can be represented by the inputs_ids passed when calling MoshiDepthDecoder. int, optional, defaults to 1024) — Dimensionality of the layers and the pooler layer of the depth decoder. int, optional, defaults to 4096) — Dimensionality of the input hidden states. Used to connect the main decoder to the depth decoder. int, optional, defaults to 6) — Number of depth decoder layers. int, optional, defaults to 16) — Number of attention heads for each attention layer in the depth decoder block. int, optional) — This is the number of key_value heads that should be used to implement Grouped Query Attention. If num_key_value_heads=num_attention_heads, the model will use Multi Head Attention (MHA), if num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed by meanpooling all the original heads within that group. For more details checkout this paper. If it is not specified, will default to num_attention_heads. int, optional, defaults to 2048) — Vocabulary size of the audio part of model. Defines the number of different tokens that can be represented by the audio_codes passed when calling the Moshi models. int, optional, defaults to 9) — The maximum sequence length that this model might ever be used with. Typically, set this to something large just in case (e.g., 512 or 1024 or 2048). str or function, optional, defaults to "silu") — The non-linear activation function (function or string) in the depth decoder. int, optional, defaults to hidden_size // num_attention_heads) — The attention head dimension. float, optional, defaults to 0.02) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices. bool, optional, defaults to True) — Whether or not the model should return the last key/values attentions (not used by all models). Only relevant if config.is_decoder=True. int, optional, defaults to 8) — Sliding window attention window size. If not specified, will default to 8. float, optional, defaults to 0.0) — The dropout ratio for the attention probabilities. int, optional, defaults to 5632) — Dimensionality of the “intermediate” (often named feed-forward) layer in the depth decoder block. Must be even. float, optional, defaults to 1e-08) — The epsilon used by the rms normalization layers. int, optional, defaults to 8) — The number of audio codebooks for each audio channels. bool, optional, defaults to False) — Whether to tie weight embeddings This is the configuration class to store the configuration of a MoshiDepthDecoder. It is used to instantiate a Moshi depth decoder model according to the specified arguments, defining the Moshi depth decoder config.
Configuration objects inherit from PretrainedConfig and can be used to control the model outputs. Read the documentation from PretrainedConfig for more information.
Example:
>>> from transformers import ( ... MoshiDepthConfig, ... MoshiDepthDecoder, ... ) >>> configuration = MoshiDepthConfig() >>> >>> model = MoshiDepthDecoder(configuration) >>> >>> configuration = model.configMoshiModel class transformers.MoshiModel < source >
( config: MoshiConfig )
Parameters
The bare Moshi Model outputting raw hidden-states without any specific head on top. This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)
This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
Transformer decoder consisting of config.num_hidden_layers layers. Each layer is a MoshiDecoderLayer
( input_ids: typing.Optional[torch.LongTensor] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Union[transformers.cache_utils.Cache, typing.List[torch.FloatTensor], NoneType] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None use_cache: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None cache_position: typing.Optional[torch.LongTensor] = None )
Parameters
torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide it.
Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
torch.Tensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:
Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
If past_key_values is used, optionally only the last input_ids have to be input (see past_key_values).
If you want to change padding behavior, you should read modeling_opt._prepare_decoder_attention_mask and modify to your needs. See diagram 1 in the paper for more information on the default strategy.
torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.n_positions - 1].
Cache or tuple(tuple(torch.FloatTensor)), optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in the past_key_values returned by the model at a previous stage of decoding, when use_cache=True or config.use_cache=True.
Two formats are allowed:
tuple(torch.FloatTensor) of length config.n_layers, with each tuple having 2 tensors of shape (batch_size, num_heads, sequence_length, embed_size_per_head)). This is also known as the legacy cache format.The model will output the same cache format that is fed as input. If no past_key_values are passed, the legacy cache format will be returned.
If past_key_values are used, the user can optionally input only the last input_ids (those that don’t have their past key value states given to this model) of shape (batch_size, 1) instead of all input_ids of shape (batch_size, sequence_length).
torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix. bool, optional) — If set to True, past_key_values key value states are returned and can be used to speed up decoding (see past_key_values). bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail. bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail. bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple. torch.LongTensor of shape (sequence_length), optional) — Indices depicting the position of the input sequence tokens in the sequence. Contrarily to position_ids, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length. The MoshiModel forward method, overrides the __call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.
( config )
Parameters
The Moshi decoder model with a text language modelling head on top. Only usable for text. This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)
This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
forward < source >( input_ids: typing.Optional[torch.LongTensor] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.LongTensor] = None past_key_values: typing.Union[transformers.cache_utils.Cache, typing.List[torch.FloatTensor], NoneType] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None use_cache: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None cache_position: typing.Optional[torch.LongTensor] = None labels: typing.Optional[torch.LongTensor] = None logits_to_keep: typing.Union[int, torch.Tensor] = 0 **kwargs ) → transformers.models.moshi.modeling_moshi.MoshiCausalLMOutputWithPast or tuple(torch.FloatTensor)
Parameters
torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide it.
Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
torch.Tensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:
Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
If past_key_values is used, optionally only the last input_ids have to be input (see past_key_values).
If you want to change padding behavior, you should read modeling_opt._prepare_decoder_attention_mask and modify to your needs. See diagram 1 in the paper for more information on the default strategy.
torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.n_positions - 1].
Cache or tuple(tuple(torch.FloatTensor)), optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in the past_key_values returned by the model at a previous stage of decoding, when use_cache=True or config.use_cache=True.
Two formats are allowed:
tuple(torch.FloatTensor) of length config.n_layers, with each tuple having 2 tensors of shape (batch_size, num_heads, sequence_length, embed_size_per_head)). This is also known as the legacy cache format.The model will output the same cache format that is fed as input. If no past_key_values are passed, the legacy cache format will be returned.
If past_key_values are used, the user can optionally input only the last input_ids (those that don’t have their past key value states given to this model) of shape (batch_size, 1) instead of all input_ids of shape (batch_size, sequence_length).
torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix. bool, optional) — If set to True, past_key_values key value states are returned and can be used to speed up decoding (see past_key_values). bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail. bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail. bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple. torch.LongTensor of shape (sequence_length), optional) — Indices depicting the position of the input sequence tokens in the sequence. Contrarily to position_ids, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length. torch.LongTensor of shape (batch_size, sequence_length), optional) — Labels for computing the masked language modeling loss. Indices should either be in [0, ..., config.vocab_size] or -100 (see input_ids docstring). Tokens with indices set to -100 are ignored (masked), the loss is only computed for the tokens with labels in [0, ..., config.vocab_size]. int or torch.Tensor, optional) — If an int, compute logits for the last logits_to_keep tokens. If 0, calculate logits for all input_ids (special case). Only last token logits are needed for generation, and calculating them only for that token can save memory, which becomes pretty significant for long sequences or large vocabulary size. If a torch.Tensor, must be 1D corresponding to the indices to keep in the sequence length dimension. This is useful when using packed tensor format (single dimension for batch and sequence length). Returns
transformers.models.moshi.modeling_moshi.MoshiCausalLMOutputWithPast or tuple(torch.FloatTensor)
A transformers.models.moshi.modeling_moshi.MoshiCausalLMOutputWithPast or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (MoshiConfig) and inputs.
loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Language modeling loss (for next-token prediction).
logits (torch.FloatTensor of shape (batch_size, sequence_length, config.vocab_size)) — Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.
past_key_values (tuple(tuple(torch.FloatTensor)), optional, returned when use_cache=True is passed or when config.use_cache=True) — Tuple of tuple(torch.FloatTensor) of length config.n_layers, with each tuple having 2 tensors of shape (batch_size, num_heads, sequence_length, embed_size_per_head))
Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see past_key_values input) to speed up sequential decoding.
hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).
Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).
Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
The MoshiForCausalLM forward method, overrides the __call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.
Example:
>>> from transformers import AutoTokenizer, MoshiForCausalLM
>>> model = MoshiForCausalLM.from_pretrained("kmhf/hf-moshiko")
>>> tokenizer = AutoTokenizer.from_pretrained("kmhf/hf-moshiko")
>>> prompt = "What is your favorite condiment?"
>>> inputs = tokenizer(prompt, return_tensors="pt")
>>>
>>> generate_ids = model.generate(inputs.input_ids, max_length=30)
>>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
"What is your favorite condiment?" MoshiForConditionalGeneration class transformers.MoshiForConditionalGeneration < source >
( config: MoshiConfig )
Parameters
The original Moshi model with an audio encoder, a Moshi depth decoder and a Moshi decoder, for speech-to-speech. This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)
This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
forward < source >( input_ids: typing.Optional[torch.LongTensor] = None attention_mask: typing.Optional[torch.BoolTensor] = None user_input_values: typing.Optional[torch.FloatTensor] = None user_audio_codes: typing.Optional[torch.Tensor] = None moshi_input_values: typing.Optional[torch.FloatTensor] = None moshi_audio_codes: typing.Optional[torch.Tensor] = None past_key_values: typing.Tuple[typing.Tuple[torch.FloatTensor]] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None text_labels: typing.Optional[torch.LongTensor] = None audio_labels: typing.Optional[torch.LongTensor] = None use_cache: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None **kwargs ) → transformers.modeling_outputs.Seq2SeqLMOutput or tuple(torch.FloatTensor)
Parameters
torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence text tokens in the vocabulary. Padding will be ignored by default should you provide it.
Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
torch.Tensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:
torch.Tensor of shape `(batch_size, 1, audio_sequence_length), optional) — The audio waveforms used as audio user prompt for the generation. torch.Tensor of shape (batch_size, num_codebooks, sequence_length), *optional*) -- The audio codes used as audio user prompt for the generation. Has priority over user_input_valuesand represents the audio "tokens" ofuser_input_values` once passed through the audio encoder. torch.Tensor of shape `(batch_size, 1, audio_sequence_length), optional) — The audio waveforms used as audio Moshi prompt for the generation. torch.Tensor of shape (batch_size, num_codebooks, sequence_length), *optional*) -- The audio codes used as audio Moshi prompt for the generation. Has priority over moshi_input_valuesand represents the audio "tokens" ofmoshi_input_values` once passed through the audio encoder. torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. If past_key_values is used, optionally only the last inputs_embeds have to be input (see past_key_values). This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
If input_ids and inputs_embeds are both unset, inputs_embeds takes the value of inputs_embeds.
Cache or tuple(tuple(torch.FloatTensor)), optional) — Pre-computed hidden-states (key and values in the self-attention blocks) that can be used to speed up sequential decoding. This typically consists in the past_key_values returned by the model at a previous stage of decoding, when use_cache=True or config.use_cache=True.
Two formats are allowed:
tuple(torch.FloatTensor) of length config.n_layers, with each tuple having 2 tensors of shape (batch_size, num_heads, sequence_length, embed_size_per_head)). This is also known as the legacy cache format.The model will output the same cache format that is fed as input. If no past_key_values are passed, the legacy cache format will be returned.
torch.LongTensor of shape (batch_size, sequence_length), optional) — Labels for text language modeling. Note that the labels are shifted inside the model, i.e. you can set labels = input_ids Indices are selected in [-100, 0, ..., config.vocab_size] All labels set to -100 are ignored (masked), the loss is only computed for labels in [0, ..., config.vocab_size] torch.LongTensor of shape (batch_size, num_codebooks, sequence_length), optional) — Labels for language modeling. Note that the labels are shifted inside the model, i.e. you can set labels = input_ids Indices are selected in [-100, 0, ..., config.vocab_size] All labels set to -100 are ignored (masked), the loss is only computed for labels in [0, ..., config.audio_vocab_size] bool, optional) — If set to True, past_key_values key value states are returned and can be used to speed up decoding (see past_key_values). bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail. bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail. bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple. A transformers.modeling_outputs.Seq2SeqLMOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (MoshiConfig) and inputs.
loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Language modeling loss.
logits (torch.FloatTensor of shape (batch_size, sequence_length, config.vocab_size)) — Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
past_key_values (tuple(tuple(torch.FloatTensor)), optional, returned when use_cache=True is passed or when config.use_cache=True) — Tuple of tuple(torch.FloatTensor) of length config.n_layers, with each tuple having 2 tensors of shape (batch_size, num_heads, sequence_length, embed_size_per_head)) and 2 additional tensors of shape (batch_size, num_heads, encoder_sequence_length, embed_size_per_head).
Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used (see past_key_values input) to speed up sequential decoding.
decoder_hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).
Hidden-states of the decoder at the output of each layer plus the initial embedding outputs.
decoder_attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).
Attentions weights of the decoder, after the attention softmax, used to compute the weighted average in the self-attention heads.
cross_attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).
Attentions weights of the decoder’s cross-attention layer, after the attention softmax, used to compute the weighted average in the cross-attention heads.
encoder_last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Sequence of hidden-states at the output of the last layer of the encoder of the model.
encoder_hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).
Hidden-states of the encoder at the output of each layer plus the initial embedding outputs.
encoder_attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).
Attentions weights of the encoder, after the attention softmax, used to compute the weighted average in the self-attention heads.
The MoshiForConditionalGeneration forward method, overrides the __call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.
Examples:
>>> from transformers import MoshiForConditionalGeneration
>>> import torch
>>> model = MoshiForConditionalGeneration.from_pretrained("kmhf/hf-moshiko")
>>> inputs = moshi.get_unconditional_inputs()
>>> logits = model(**inputs, ).logits
>>> logits.shape
torch.Size([1, 1, 32000]) generate < source >
( input_ids: typing.Optional[torch.LongTensor] = None user_input_values: typing.Optional[torch.FloatTensor] = None user_audio_codes: typing.Optional[torch.Tensor] = None moshi_input_values: typing.Optional[torch.FloatTensor] = None moshi_audio_codes: typing.Optional[torch.Tensor] = None inputs_embeds: typing.Optional[torch.FloatTensor] = None return_audio_waveforms: typing.Optional[bool] = True return_audio_codes: typing.Optional[bool] = None concat_unconditional_inputs: typing.Optional[bool] = True **kwargs )
Parameters
torch.Tensor of shape `(batch_size, sequence_length), optional) — The sequence used as a text prompt for the generation. torch.Tensor of shape `(batch_size, 1, audio_sequence_length), optional) — The audio waveforms used as audio user prompt for the generation. torch.Tensor of shape (batch_size, num_codebooks, sequence_length), *optional*) -- The audio codes used as audio user prompt for the generation. Has priority over user_input_valuesand represents the audio "tokens" ofuser_input_values` once passed through the audio encoder. torch.Tensor of shape `(batch_size, 1, audio_sequence_length), optional) — The audio waveforms used as audio Moshi prompt for the generation. torch.Tensor of shape (batch_size, num_codebooks, sequence_length), *optional*) -- The audio codes used as audio Moshi prompt for the generation. Has priority over moshi_input_valuesand represents the audio "tokens" ofmoshi_input_values` once passed through the audio encoder. torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids and the audio inputs you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert the inputs into associated vectors than the model’s internal embedding lookup matrix. bool, optional, defaults to True) — If False, won’t generate the audio waveforms. bool, optional) — If True, will also returns the generated audio codes, i.e the intermediate audio “tokens” which transforms to audio_sequences once passed through the audio decoder. bool, optional, defaults to True) — If False, won’t concatenate initial audio and text tokens. Dict[str, Any], optional) — Remaining dictionary of keyword arguments that are passed to the generate method. Refers to the original generate docstrings for more information on how to use them. Note that keywords with a depth_ prefix will be input for the generate method of the depth decoder. Otherwise, the latter will use its default generation config. Generates sequences of text token ids and audio tokens ids.
get_unconditional_inputs < source >( num_samples = 1 )
Parameters
Helper function to get null inputs for unconditional generation, enabling the model to be used without the feature extractor or tokenizer.
Example:
>>> from transformers import MoshiForConditionalGeneration
>>> model = MoshiForConditionalGeneration.from_pretrained("kmhf/hf-moshiko-pytorch-bf16")
>>>
>>> unconditional_inputs = model.get_unconditional_inputs(num_samples=1)
>>> audio_samples = model.generate(**unconditional_inputs, max_new_tokens=256)< > Update on GitHub
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.3