The Speech2Text2 model is used together with Wav2Vec2 for Speech Translation models proposed in Large-Scale Self- and Semi-Supervised Learning for Speech Translation by Changhan Wang, Anne Wu, Juan Pino, Alexei Baevski, Michael Auli, Alexis Conneau.
Speech2Text2 is a decoder-only transformer model that can be used with any speech encoder-only, such as Wav2Vec2 or HuBERT for Speech-to-Text tasks. Please refer to the SpeechEncoderDecoder class on how to combine Speech2Text2 with any speech encoder-only model.
This model was contributed by Patrick von Platen.
The original code can be found here.
Speech2Text2’s SpeechEncoderDecoderModel model accepts raw waveform input values from speech and makes use of generate() to translate the input speech autoregressively to the target language.
The Wav2Vec2FeatureExtractor class is responsible for preprocessing the input speech and Speech2Text2Tokenizer decodes the generated target tokens to the target string. The Speech2Text2Processor wraps Wav2Vec2FeatureExtractor and Speech2Text2Tokenizer into a single instance to both extract the input features and decode the predicted token ids.
>>> import torch
>>> from transformers import Speech2Text2Processor, SpeechEncoderDecoderModel
>>> from datasets import load_dataset
>>> import soundfile as sf
>>> model = SpeechEncoderDecoderModel.from_pretrained("facebook/s2t-wav2vec2-large-en-de")
>>> processor = Speech2Text2Processor.from_pretrained("facebook/s2t-wav2vec2-large-en-de")
>>> def map_to_array(batch):
... speech, _ = sf.read(batch["file"])
... batch["speech"] = speech
... return batch
>>> ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
>>> ds = ds.map(map_to_array)
>>> inputs = processor(ds["speech"][0], sampling_rate=16_000, return_tensors="pt")
>>> generated_ids = model.generate(inputs=inputs["input_values"], attention_mask=inputs["attention_mask"])
>>> transcription = processor.batch_decode(generated_ids)Speech Translation via Pipelines
The automatic speech recognition pipeline can also be used to translate speech in just a couple lines of code
>>> from datasets import load_dataset
>>> from transformers import pipeline
>>> librispeech_en = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
>>> asr = pipeline(
... "automatic-speech-recognition",
... model="facebook/s2t-wav2vec2-large-en-de",
... feature_extractor="facebook/s2t-wav2vec2-large-en-de",
... )
>>> translation_de = asr(librispeech_en[0]["file"])See model hub to look for Speech2Text2 checkpoints.
( vocab_size = 10000 decoder_layers = 6 decoder_ffn_dim = 2048 decoder_attention_heads = 4 decoder_layerdrop = 0.0 use_cache = True activation_function = 'relu' d_model = 256 dropout = 0.1 attention_dropout = 0.0 activation_dropout = 0.0 init_std = 0.02 decoder_start_token_id = 2 scale_embedding = True pad_token_id = 1 bos_token_id = 0 eos_token_id = 2 max_target_positions = 1024 **kwargs )
Parameters
int, optional, defaults to 50265) —
Vocabulary size of the Speech2Text model. Defines the number of different tokens that can be represented by
the inputs_ids passed when calling Speech2TextModel int, optional, defaults to 1024) —
Dimensionality of the layers and the pooler layer. int, optional, defaults to 12) —
Number of decoder layers. int, optional, defaults to 16) —
Number of attention heads for each attention layer in the Transformer decoder. int, optional, defaults to 4096) —
Dimensionality of the “intermediate” (often named feed-forward) layer in decoder. str or function, optional, defaults to "gelu") —
The non-linear activation function (function or string) in the pooler. If string, "gelu", "relu",
"silu" and "gelu_new" are supported. float, optional, defaults to 0.1) —
The dropout probability for all fully connected layers in the embeddings, and pooler. float, optional, defaults to 0.0) —
The dropout ratio for the attention probabilities. float, optional, defaults to 0.0) —
The dropout ratio for activations inside the fully connected layer. float, optional, defaults to 0.02) —
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
https://arxiv.org/abs/1909.11556>`__ for more details. float, optional, defaults to 0.0) —
The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556)
for more details. bool, optional, defaults to True) —
Whether or not the model should return the last key/values attentions (not used by all models). int, optional, defaults to 1024) —
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). This is the configuration class to store the configuration of a Speech2Text2ForCausalLM. It is used to instantiate an Speech2Text2 model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the Speech2Text2 facebook/s2t-wav2vec2-large-en-de architecture.
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 Speech2Text2Config, Speech2Text2ForCausalLM
>>> # Initializing a Speech2Text2 s2t_transformer_s style configuration
>>> configuration = Speech2Text2Config()
>>> # Initializing a model (with random weights) from the s2t_transformer_s style configuration
>>> model = Speech2Text2ForCausalLM(configuration)
>>> # Accessing the model configuration
>>> configuration = model.config( vocab_file bos_token = '<s>' pad_token = '<pad>' eos_token = '</s>' unk_token = '<unk>' do_lower_case = False merges_file = None **kwargs )
Parameters
str) —
File containing the vocabulary. str, optional, defaults to "<s>") —
The beginning of sentence token. str, optional, defaults to "</s>") —
The end of sentence token. str, optional, defaults to "<unk>") —
The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this
token instead. str, optional, defaults to "<pad>") —
The token used for padding, for example when batching sequences of different lengths.
**kwargs — Additional keyword arguments passed along to PreTrainedTokenizer
Constructs a Speech2Text2Tokenizer.
This tokenizer inherits from PreTrainedTokenizer which contains some of the main methods. Users should refer to the superclass for more information regarding such methods.
( sequences: Union skip_special_tokens: bool = False clean_up_tokenization_spaces: bool = None **kwargs ) → List[str]
Parameters
Union[List[int], List[List[int]], np.ndarray, torch.Tensor, tf.Tensor]) —
List of tokenized input ids. Can be obtained using the __call__ method. bool, optional, defaults to False) —
Whether or not to remove special tokens in the decoding. bool, optional) —
Whether or not to clean up the tokenization spaces. If None, will default to
self.clean_up_tokenization_spaces. Returns
List[str]
The list of decoded sentences.
Convert a list of lists of token ids into a list of strings by calling decode.
( token_ids: Union skip_special_tokens: bool = False clean_up_tokenization_spaces: bool = None **kwargs ) → str
Parameters
Union[int, List[int], np.ndarray, torch.Tensor, tf.Tensor]) —
List of tokenized input ids. Can be obtained using the __call__ method. bool, optional, defaults to False) —
Whether or not to remove special tokens in the decoding. bool, optional) —
Whether or not to clean up the tokenization spaces. If None, will default to
self.clean_up_tokenization_spaces. Returns
str
The decoded sentence.
Converts a sequence of ids in a string, using the tokenizer and vocabulary with options to remove special tokens and clean up tokenization spaces.
Similar to doing self.convert_tokens_to_string(self.convert_ids_to_tokens(token_ids)).
( feature_extractor tokenizer )
Parameters
AutoFeatureExtractor) —
An instance of AutoFeatureExtractor. The feature extractor is a required input. Speech2Text2Tokenizer) —
An instance of Speech2Text2Tokenizer. The tokenizer is a required input. Constructs a Speech2Text2 processor which wraps a Speech2Text2 feature extractor and a Speech2Text2 tokenizer into a single processor.
Speech2Text2Processor offers all the functionalities of AutoFeatureExtractor and Speech2Text2Tokenizer. See the call() and decode() for more information.
When used in normal mode, this method forwards all its arguments to AutoFeatureExtractor’s
__call__() and returns its output. If used in the context
as_target_processor() this method forwards all its arguments to
Speech2Text2Tokenizer’s call(). Please refer to the doctsring of the above two
methods for more information.
( pretrained_model_name_or_path: Union cache_dir: Union = None force_download: bool = False local_files_only: bool = False token: Union = None revision: str = 'main' **kwargs )
Parameters
str or os.PathLike) —
This can be either:
./my_model_directory/../my_model_directory/preprocessor_config.json.
**kwargs —
Additional keyword arguments passed along to both
from_pretrained() and
~tokenization_utils_base.PreTrainedTokenizer.from_pretrained.Instantiate a processor associated with a pretrained model.
This class method is simply calling the feature extractor
from_pretrained(), image processor
ImageProcessingMixin and the tokenizer
~tokenization_utils_base.PreTrainedTokenizer.from_pretrained methods. Please refer to the docstrings of the
methods above for more information.
( save_directory push_to_hub: bool = False **kwargs )
Parameters
str or os.PathLike) —
Directory where the feature extractor JSON file and the tokenizer files will be saved (directory will
be created if it does not exist). bool, optional, defaults to False) —
Whether or not to push your model to the Hugging Face model hub after saving it. You can specify the
repository you want to push to with repo_id (will default to the name of save_directory in your
namespace). Dict[str, Any], optional) —
Additional key word arguments passed along to the push_to_hub() method. Saves the attributes of this processor (feature extractor, tokenizer…) in the specified directory so that it can be reloaded using the from_pretrained() method.
This class method is simply calling save_pretrained() and save_pretrained(). Please refer to the docstrings of the methods above for more information.
This method forwards all its arguments to Speech2Text2Tokenizer’s batch_decode(). Please refer to the docstring of this method for more information.
This method forwards all its arguments to Speech2Text2Tokenizer’s decode(). Please refer to the docstring of this method for more information.
( config )
Parameters
The Speech2Text2 Decoder with a language modeling head. Can be used as the decoder part of EncoderDecoderModel and SpeechEncoderDecoder.
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.
( input_ids: Optional = None attention_mask: Optional = None encoder_hidden_states: Optional = None encoder_attention_mask: Optional = None head_mask: Optional = None cross_attn_head_mask: Optional = None past_key_values: Optional = None inputs_embeds: Optional = None labels: Optional = None use_cache: Optional = None output_attentions: Optional = None output_hidden_states: Optional = None return_dict: Optional = None ) → transformers.modeling_outputs.CausalLMOutputWithCrossAttentions 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 Speech2Text2Tokenizer. 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.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) —
Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention
if the model is configured as a decoder. torch.FloatTensor of shape (batch_size, sequence_length), optional) —
Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used
in the cross-attention if the model is configured as a decoder. Mask values selected in [0, 1]: torch.Tensor of shape (decoder_layers, decoder_attention_heads), optional) —
Mask to nullify selected heads of the attention modules. Mask values selected in [0, 1]:
torch.Tensor of shape (decoder_layers, decoder_attention_heads), optional) —
Mask to nullify selected heads of the cross-attention modules. Mask values selected in [0, 1]:
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). The two additional
tensors are only required when the model is used as a decoder in a Sequence to Sequence model.
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.
If past_key_values are used, the user can optionally input only the last decoder_input_ids (those
that don’t have their past key value states given to this model) of shape (batch_size, 1) instead of
all decoder_input_ids of shape (batch_size, 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]. 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. Returns
transformers.modeling_outputs.CausalLMOutputWithCrossAttentions or tuple(torch.FloatTensor)
A transformers.modeling_outputs.CausalLMOutputWithCrossAttentions 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 (Speech2Text2Config) 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).
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.
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).
Cross attentions weights after the attention softmax, used to compute the weighted average in the cross-attention heads.
past_key_values (tuple(tuple(torch.FloatTensor)), optional, returned when use_cache=True is passed or when config.use_cache=True) — Tuple of torch.FloatTensor tuples of length config.n_layers, with each tuple containing the cached key,
value states of the self-attention and the cross-attention layers if model is used in encoder-decoder
setting. Only relevant if config.is_decoder = True.
Contains pre-computed hidden-states (key and values in the attention blocks) that can be used (see
past_key_values input) to speed up sequential decoding.
Example:
>>> from transformers import (
... SpeechEncoderDecoderModel,
... Speech2Text2ForCausalLM,
... Wav2Vec2Model,
... Speech2Text2Config,
... Wav2Vec2Config,
... Wav2Vec2FeatureExtractor,
... Speech2Text2Tokenizer,
... )
>>> from datasets import load_dataset
>>> feature_extractor = Wav2Vec2FeatureExtractor()
>>> tokenizer = Speech2Text2Tokenizer.from_pretrained("facebook/s2t-wav2vec2-large-en-de")
>>> encoder = Wav2Vec2Model(Wav2Vec2Config())
>>> decoder = Speech2Text2ForCausalLM(Speech2Text2Config())
>>> # init random speech2text model
>>> model = SpeechEncoderDecoderModel(encoder=encoder, decoder=decoder)
>>> model.config.pad_token_id = tokenizer.pad_token_id
>>> model.config.decoder_start_token_id = tokenizer.bos_token_id
>>> # pre-process inputs and labels
>>> ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
>>> inputs = feature_extractor(
... ds[0]["audio"]["array"], sampling_rate=ds[0]["audio"]["sampling_rate"], return_tensors="pt"
... )
>>> input_values = inputs.input_values
>>> decoder_input_ids = tokenizer(ds[0]["text"], return_tensors="pt").input_ids
>>> # compute loss
>>> loss = model(inputs=input_values, labels=decoder_input_ids).loss
>>> # backprop loss
>>> loss.backward()