The Wav2Vec2-BERT model was proposed in Seamless: Multilingual Expressive and Streaming Speech Translation by the Seamless Communication team from Meta AI.
This model was pre-trained on 4.5M hours of unlabeled audio data covering more than 143 languages. It requires finetuning to be used for downstream tasks such as Automatic Speech Recognition (ASR), or Audio Classification.
The official results of the model can be found in Section 3.2.1 of the paper.
The abstract from the paper is the following:
Recent advancements in automatic speech translation have dramatically expanded language coverage, improved multimodal capabilities, and enabled a wide range of tasks and functionalities. That said, large-scale automatic speech translation systems today lack key features that help machine-mediated communication feel seamless when compared to human-to-human dialogue. In this work, we introduce a family of models that enable end-to-end expressive and multilingual translations in a streaming fashion. First, we contribute an improved version of the massively multilingual and multimodal SeamlessM4T model—SeamlessM4T v2. This newer model, incorporating an updated UnitY2 framework, was trained on more low-resource language data. The expanded version of SeamlessAlign adds 114,800 hours of automatically aligned data for a total of 76 languages. SeamlessM4T v2 provides the foundation on which our two newest models, SeamlessExpressive and SeamlessStreaming, are initiated. SeamlessExpressive enables translation that preserves vocal styles and prosody. Compared to previous efforts in expressive speech research, our work addresses certain underexplored aspects of prosody, such as speech rate and pauses, while also preserving the style of one’s voice. As for SeamlessStreaming, our model leverages the Efficient Monotonic Multihead Attention (EMMA) mechanism to generate low-latency target translations without waiting for complete source utterances. As the first of its kind, SeamlessStreaming enables simultaneous speech-to-speech/text translation for multiple source and target languages. To understand the performance of these models, we combined novel and modified versions of existing automatic metrics to evaluate prosody, latency, and robustness. For human evaluations, we adapted existing protocols tailored for measuring the most relevant attributes in the preservation of meaning, naturalness, and expressivity. To ensure that our models can be used safely and responsibly, we implemented the first known red-teaming effort for multimodal machine translation, a system for the detection and mitigation of added toxicity, a systematic evaluation of gender bias, and an inaudible localized watermarking mechanism designed to dampen the impact of deepfakes. Consequently, we bring major components from SeamlessExpressive and SeamlessStreaming together to form Seamless, the first publicly available system that unlocks expressive cross-lingual communication in real-time. In sum, Seamless gives us a pivotal look at the technical foundation needed to turn the Universal Speech Translator from a science fiction concept into a real-world technology. Finally, contributions in this work—including models, code, and a watermark detector—are publicly released and accessible at the link below.
This model was contributed by ylacombe. The original code can be found here.
config.position_embeddings_type.( vocab_size = None hidden_size = 1024 num_hidden_layers = 24 num_attention_heads = 16 intermediate_size = 4096 feature_projection_input_dim = 160 hidden_act = 'swish' hidden_dropout = 0.0 activation_dropout = 0.0 attention_dropout = 0.0 feat_proj_dropout = 0.0 final_dropout = 0.1 layerdrop = 0.1 initializer_range = 0.02 layer_norm_eps = 1e-05 apply_spec_augment = True mask_time_prob = 0.05 mask_time_length = 10 mask_time_min_masks = 2 mask_feature_prob = 0.0 mask_feature_length = 10 mask_feature_min_masks = 0 ctc_loss_reduction = 'sum' ctc_zero_infinity = False use_weighted_layer_sum = False classifier_proj_size = 768 tdnn_dim = (512, 512, 512, 512, 1500) tdnn_kernel = (5, 3, 3, 1, 1) tdnn_dilation = (1, 2, 3, 1, 1) xvector_output_dim = 512 pad_token_id = 0 bos_token_id = 1 eos_token_id = 2 add_adapter = False adapter_kernel_size = 3 adapter_stride = 2 num_adapter_layers = 1 adapter_act = 'relu' use_intermediate_ffn_before_adapter = False output_hidden_size = None position_embeddings_type = 'relative_key' rotary_embedding_base = 10000 max_source_positions = 5000 left_max_position_embeddings = 64 right_max_position_embeddings = 8 conv_depthwise_kernel_size = 31 conformer_conv_dropout = 0.1 **kwargs )
Parameters
int, optional) —
Vocabulary size of the Wav2Vec2Bert model. Defines the number of different tokens that can be
represented by the inputs_ids passed when calling Wav2Vec2BertModel. Vocabulary size of the
model. Defines the different tokens that can be represented by the inputs_ids passed to the forward
method of Wav2Vec2BertModel. int, optional, defaults to 1024) —
Dimensionality of the encoder layers and the pooler layer. int, optional, defaults to 24) —
Number of hidden layers in the Transformer encoder. int, optional, defaults to 16) —
Number of attention heads for each attention layer in the Transformer encoder. int, optional, defaults to 4096) —
Dimensionality of the “intermediate” (i.e., feed-forward) layer in the Transformer encoder. int, optional, defaults to 160) —
Input dimension of this model, i.e the dimension after processing input audios with SeamlessM4TFeatureExtractor or Wav2Vec2BertProcessor. str or function, optional, defaults to "swish") —
The non-linear activation function (function or string) in the encoder and pooler. If string, "gelu",
"relu", "selu", "swish" and "gelu_new" are supported. float, optional, defaults to 0.0) —
The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. float, optional, defaults to 0.0) —
The dropout ratio for activations inside the fully connected layer. float, optional, defaults to 0.0) —
The dropout ratio for the attention probabilities. float, optional, defaults to 0.0) —
The dropout probability for the feature projection. float, optional, defaults to 0.1) —
The dropout probability for the final projection layer of Wav2Vec2BertForCTC. float, optional, defaults to 0.1) —
The LayerDrop probability. See the [LayerDrop paper](see https://arxiv.org/abs/1909.11556) for more
details. float, optional, defaults to 0.02) —
The standard deviation of the truncated_normal_initializer for initializing all weight matrices. float, optional, defaults to 1e-05) —
The epsilon used by the layer normalization layers. bool, optional, defaults to True) —
Whether to apply SpecAugment data augmentation to the outputs of the feature encoder. For reference see
SpecAugment: A Simple Data Augmentation Method for Automatic Speech
Recognition. float, optional, defaults to 0.05) —
Percentage (between 0 and 1) of all feature vectors along the time axis which will be masked. The masking
procecure generates mask_time_prob*len(time_axis)/mask_time_length ``independent masks over the axis. If reasoning from the propability of each feature vector to be chosen as the start of the vector span to be masked, *mask_time_prob* should be prob_vector_start*mask_time_length. Note that overlap may decrease the actual percentage of masked vectors. This is only relevant if apply_spec_augment is True`. int, optional, defaults to 10) —
Length of vector span along the time axis. int, optional, defaults to 2) —
The minimum number of masks of length mask_feature_length generated along the time axis, each time step,
irrespectively of mask_feature_prob. Only relevant if mask_time_prob*len(time_axis)/mask_time_length < mask_time_min_masks. float, optional, defaults to 0.0) —
Percentage (between 0 and 1) of all feature vectors along the feature axis which will be masked. The
masking procecure generates mask_feature_prob*len(feature_axis)/mask_time_length independent masks over
the axis. If reasoning from the propability of each feature vector to be chosen as the start of the vector
span to be masked, mask_feature_prob should be prob_vector_start*mask_feature_length. Note that overlap
may decrease the actual percentage of masked vectors. This is only relevant if apply_spec_augment is True. int, optional, defaults to 10) —
Length of vector span along the feature axis. int, optional, defaults to 0) —
The minimum number of masks of length mask_feature_length generated along the feature axis, each time
step, irrespectively of mask_feature_prob. Only relevant if
mask_feature_prob*len(feature_axis)/mask_feature_length < mask_feature_min_masks. str, optional, defaults to "sum") —
Specifies the reduction to apply to the output of torch.nn.CTCLoss. Only relevant when training an
instance of Wav2Vec2BertForCTC. bool, optional, defaults to False) —
Whether to zero infinite losses and the associated gradients of torch.nn.CTCLoss. Infinite losses mainly
occur when the inputs are too short to be aligned to the targets. Only relevant when training an instance
of Wav2Vec2BertForCTC. bool, optional, defaults to False) —
Whether to use a weighted average of layer outputs with learned weights. Only relevant when using an
instance of Wav2Vec2BertForSequenceClassification. int, optional, defaults to 768) —
Dimensionality of the projection before token mean-pooling for classification. Tuple[int] or List[int], optional, defaults to (512, 512, 512, 512, 1500)) —
A tuple of integers defining the number of output channels of each 1D convolutional layer in the TDNN
module of the XVector model. The length of tdnn_dim defines the number of TDNN layers. Tuple[int] or List[int], optional, defaults to (5, 3, 3, 1, 1)) —
A tuple of integers defining the kernel size of each 1D convolutional layer in the TDNN module of the
XVector model. The length of tdnn_kernel has to match the length of tdnn_dim. Tuple[int] or List[int], optional, defaults to (1, 2, 3, 1, 1)) —
A tuple of integers defining the dilation factor of each 1D convolutional layer in TDNN module of the
XVector model. The length of tdnn_dilation has to match the length of tdnn_dim. int, optional, defaults to 512) —
Dimensionality of the XVector embedding vectors. int, optional, defaults to 0) — The id of the beginning-of-stream token. int, optional, defaults to 1) — The id of the padding token. int, optional, defaults to 2) — The id of the end-of-stream token. bool, optional, defaults to False) —
Whether a convolutional attention network should be stacked on top of the Wav2Vec2Bert Encoder. Can be very
useful for warm-starting Wav2Vec2Bert for SpeechEncoderDecoder models. int, optional, defaults to 3) —
Kernel size of the convolutional layers in the adapter network. Only relevant if add_adapter is True. int, optional, defaults to 2) —
Stride of the convolutional layers in the adapter network. Only relevant if add_adapter is True. int, optional, defaults to 1) —
Number of convolutional layers that should be used in the adapter network. Only relevant if add_adapter is True. str or function, optional, defaults to "relu") —
The non-linear activation function (function or string) in the adapter layers. If string, "gelu",
"relu", "selu", "swish" and "gelu_new" are supported. bool, optional, defaults to False) —
Whether an intermediate feed-forward block should be stacked on top of the Wav2Vec2Bert Encoder and before the adapter network.
Only relevant if add_adapter is True. int, optional) —
Dimensionality of the encoder output layer. If not defined, this defaults to hidden-size. Only relevant
if add_adapter is True. str, optional, defaults to "relative_key") —
Can be specified to :rotary, for rotary position embeddings.relative, for relative position embeddings.relative_key, for relative position embeddings as defined by Shaw in Self-Attention
with Relative Position Representations (Shaw et al.).
If left to None, no relative position embeddings is applied.int, optional, defaults to 10000) —
If "rotary" position embeddings are used, defines the size of the embedding base. int, optional, defaults to 5000) —
if "relative" position embeddings are used, defines the maximum source input positions. int, optional, defaults to 64) —
If "relative_key" (aka Shaw) position embeddings are used, defines the left clipping value for relative positions. int, optional, defaults to 8) —
If "relative_key" (aka Shaw) position embeddings are used, defines the right clipping value for relative positions. int, optional, defaults to 31) —
Kernel size of convolutional depthwise 1D layer in Conformer blocks. float, optional, defaults to 0.1) —
The dropout probability for all convolutional layers in Conformer blocks. This is the configuration class to store the configuration of a Wav2Vec2BertModel. It is used to instantiate an Wav2Vec2Bert 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 Wav2Vec2Bert facebook/wav2vec2-bert-rel-pos-large 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 Wav2Vec2BertConfig, Wav2Vec2BertModel
>>> # Initializing a Wav2Vec2Bert facebook/wav2vec2-bert-rel-pos-large style configuration
>>> configuration = Wav2Vec2BertConfig()
>>> # Initializing a model (with random weights) from the facebook/wav2vec2-bert-rel-pos-large style configuration
>>> model = Wav2Vec2BertModel(configuration)
>>> # Accessing the model configuration
>>> configuration = model.config( feature_extractor tokenizer )
Parameters
SeamlessM4TFeatureExtractor) —
An instance of SeamlessM4TFeatureExtractor. The feature extractor is a required input. Constructs a Wav2Vec2-BERT processor which wraps a Wav2Vec2-BERT feature extractor and a Wav2Vec2 CTC tokenizer into a single processor.
Wav2Vec2Processor offers all the functionalities of SeamlessM4TFeatureExtractor and PreTrainedTokenizer. See the docstring of call() and decode() for more information.
( audio = None text = None **kwargs ) → BatchEncoding
Parameters
str, List[str], List[List[str]]) —
The sequence or batch of sequences to be encoded. Each sequence can be a string or a list of strings
(pretokenized string). If the sequences are provided as list of strings (pretokenized), you must set
is_split_into_words=True (to lift the ambiguity with a batch of sequences). np.ndarray, torch.Tensor, List[np.ndarray], List[torch.Tensor]) —
The audio or batch of audios to be prepared. Each audio can be NumPy array or PyTorch tensor. In case
of a NumPy array/PyTorch tensor, each audio should be of shape (C, T), where C is a number of channels,
and T the sample length of the audio. Returns
A BatchEncoding with the following fields:
audio is not None.audio is not None.
When only text is specified, returns the token attention mask.text and audio are not None.text is not None and audio is None.Main method to prepare for the model one or several sequences(s) and audio(s). This method forwards the audio
and kwargs arguments to SeamlessM4TFeatureExtractor’s call() if audio is not
None to pre-process the audio. To prepare the target sequences(s), this method forwards the text and kwargs arguments to
PreTrainedTokenizer’s call() if text is not None. Please refer to the doctsring of the above two methods for more information.
If input_features is not None, this method forwards the input_features and kwargs arguments to SeamlessM4TFeatureExtractor’s pad() to pad the input features.
If labels is not None, this method forwards the labels and kwargs arguments to PreTrainedTokenizer’s pad() to pad the label(s).
Please refer to the doctsring of the above two methods 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 PreTrainedTokenizer’s batch_decode(). Please refer to the docstring of this method for more information.
This method forwards all its arguments to PreTrainedTokenizer’s decode(). Please refer to the docstring of this method for more information.
( config: Wav2Vec2BertConfig )
Parameters
The bare Wav2Vec2Bert Model transformer outputting raw hidden-states without any specific head on top. Wav2Vec2Bert was proposed in wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations by Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, Michael Auli.
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 etc.).
This model is a PyTorch nn.Module sub-class. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
( input_features: Optional attention_mask: Optional = None mask_time_indices: Optional = None output_attentions: Optional = None output_hidden_states: Optional = None return_dict: Optional = None ) → transformers.modeling_outputs.Wav2Vec2BaseModelOutput or tuple(torch.FloatTensor)
Parameters
torch.FloatTensor of shape (batch_size, sequence_length)) —
Float values of input raw speech waveform. Values can be obtained by loading a .flac or .wav audio file
into an array of type List[float] or a numpy.ndarray, e.g. via the soundfile library (pip install soundfile). To prepare the array into input_features, the AutoProcessor should be used for padding and
conversion into a tensor of type torch.FloatTensor. See Wav2Vec2BertProcessor.call() for details. torch.LongTensor of shape (batch_size, sequence_length), optional) —
Mask to avoid performing convolution and attention on padding token indices. Mask values selected in [0, 1]:
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.Wav2Vec2BaseModelOutput or tuple(torch.FloatTensor)
A transformers.modeling_outputs.Wav2Vec2BaseModelOutput 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 (Wav2Vec2BertConfig) and inputs.
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.
extract_features (torch.FloatTensor of shape (batch_size, sequence_length, conv_dim[-1])) — Sequence of extracted feature vectors of the last convolutional layer of the model.
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 + 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 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 Wav2Vec2BertModel 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 AutoProcessor, Wav2Vec2BertModel
>>> import torch
>>> from datasets import load_dataset
>>> dataset = load_dataset("hf-internal-testing/librispeech_asr_demo", "clean", split="validation")
>>> dataset = dataset.sort("id")
>>> sampling_rate = dataset.features["audio"].sampling_rate
>>> processor = AutoProcessor.from_pretrained("hf-audio/wav2vec2-bert-CV16-en")
>>> model = Wav2Vec2BertModel.from_pretrained("hf-audio/wav2vec2-bert-CV16-en")
>>> # audio file is decoded on the fly
>>> inputs = processor(dataset[0]["audio"]["array"], sampling_rate=sampling_rate, return_tensors="pt")
>>> with torch.no_grad():
... outputs = model(**inputs)
>>> last_hidden_states = outputs.last_hidden_state
>>> list(last_hidden_states.shape)
[1, 146, 1024]( config target_lang: Optional = None )
Parameters
Wav2Vec2Bert Model with a language modeling head on top for Connectionist Temporal Classification (CTC).
Wav2Vec2Bert was proposed in wav2vec 2.0: A Framework for Self-Supervised Learning of Speech
Representations by Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, Michael
Auli.
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 etc.).
This model is a PyTorch nn.Module sub-class. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
( input_features: Optional attention_mask: Optional = None output_attentions: Optional = None output_hidden_states: Optional = None return_dict: Optional = None labels: Optional = None ) → transformers.modeling_outputs.CausalLMOutput or tuple(torch.FloatTensor)
Parameters
torch.FloatTensor of shape (batch_size, sequence_length)) —
Float values of input raw speech waveform. Values can be obtained by loading a .flac or .wav audio file
into an array of type List[float] or a numpy.ndarray, e.g. via the soundfile library (pip install soundfile). To prepare the array into input_features, the AutoProcessor should be used for padding and
conversion into a tensor of type torch.FloatTensor. See Wav2Vec2BertProcessor.call() for details. torch.LongTensor of shape (batch_size, sequence_length), optional) —
Mask to avoid performing convolution and attention on padding token indices. Mask values selected in [0, 1]:
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 (batch_size, target_length), optional) —
Labels for connectionist temporal classification. Note that target_length has to be smaller or equal to
the sequence length of the output logits. Indices are selected in [-100, 0, ..., config.vocab_size - 1].
All labels set to -100 are ignored (masked), the loss is only computed for labels in [0, ..., config.vocab_size - 1]. Returns
transformers.modeling_outputs.CausalLMOutput or tuple(torch.FloatTensor)
A transformers.modeling_outputs.CausalLMOutput 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 (Wav2Vec2BertConfig) 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.
The Wav2Vec2BertForCTC 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 AutoProcessor, Wav2Vec2BertForCTC
>>> from datasets import load_dataset
>>> import torch
>>> dataset = load_dataset("hf-internal-testing/librispeech_asr_demo", "clean", split="validation")
>>> dataset = dataset.sort("id")
>>> sampling_rate = dataset.features["audio"].sampling_rate
>>> processor = AutoProcessor.from_pretrained("hf-audio/wav2vec2-bert-CV16-en")
>>> model = Wav2Vec2BertForCTC.from_pretrained("hf-audio/wav2vec2-bert-CV16-en")
>>> # audio file is decoded on the fly
>>> inputs = processor(dataset[0]["audio"]["array"], sampling_rate=sampling_rate, return_tensors="pt")
>>> with torch.no_grad():
... logits = model(**inputs).logits
>>> predicted_ids = torch.argmax(logits, dim=-1)
>>> # transcribe speech
>>> transcription = processor.batch_decode(predicted_ids)
>>> transcription[0]
'mr quilter is the apostle of the middle classes and we are glad to welcome his gospel'
>>> inputs["labels"] = processor(text=dataset[0]["text"], return_tensors="pt").input_ids
>>> # compute loss
>>> loss = model(**inputs).loss
>>> round(loss.item(), 2)
17.04( config )
Parameters
Wav2Vec2Bert Model with a sequence classification head on top (a linear layer over the pooled output) for tasks like SUPERB Keyword Spotting.
Wav2Vec2Bert was proposed in wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations by Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, Michael Auli.
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 etc.).
This model is a PyTorch nn.Module sub-class. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
( input_features: Optional attention_mask: Optional = None output_attentions: Optional = None output_hidden_states: Optional = None return_dict: Optional = None labels: Optional = None ) → transformers.modeling_outputs.SequenceClassifierOutput or tuple(torch.FloatTensor)
Parameters
torch.FloatTensor of shape (batch_size, sequence_length)) —
Float values of input raw speech waveform. Values can be obtained by loading a .flac or .wav audio file
into an array of type List[float] or a numpy.ndarray, e.g. via the soundfile library (pip install soundfile). To prepare the array into input_features, the AutoProcessor should be used for padding and
conversion into a tensor of type torch.FloatTensor. See Wav2Vec2BertProcessor.call() for details. torch.LongTensor of shape (batch_size, sequence_length), optional) —
Mask to avoid performing convolution and attention on padding token indices. Mask values selected in [0, 1]:
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 (batch_size,), optional) —
Labels for computing the sequence classification/regression loss. Indices should be in [0, ..., config.num_labels - 1]. If config.num_labels == 1 a regression loss is computed (Mean-Square loss), If
config.num_labels > 1 a classification loss is computed (Cross-Entropy). Returns
transformers.modeling_outputs.SequenceClassifierOutput or tuple(torch.FloatTensor)
A transformers.modeling_outputs.SequenceClassifierOutput 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 (Wav2Vec2BertConfig) and inputs.
loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Classification (or regression if config.num_labels==1) loss.
logits (torch.FloatTensor of shape (batch_size, config.num_labels)) — Classification (or regression if config.num_labels==1) scores (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.
The Wav2Vec2BertForSequenceClassification 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 AutoFeatureExtractor, Wav2Vec2BertForSequenceClassification
>>> from datasets import load_dataset
>>> import torch
>>> dataset = load_dataset("hf-internal-testing/librispeech_asr_demo", "clean", split="validation")
>>> dataset = dataset.sort("id")
>>> sampling_rate = dataset.features["audio"].sampling_rate
>>> feature_extractor = AutoFeatureExtractor.from_pretrained("facebook/w2v-bert-2.0")
>>> model = Wav2Vec2BertForSequenceClassification.from_pretrained("facebook/w2v-bert-2.0")
>>> # audio file is decoded on the fly
>>> inputs = feature_extractor(dataset[0]["audio"]["array"], sampling_rate=sampling_rate, return_tensors="pt")
>>> with torch.no_grad():
... logits = model(**inputs).logits
>>> predicted_class_ids = torch.argmax(logits, dim=-1).item()
>>> predicted_label = model.config.id2label[predicted_class_ids]
>>> # compute loss - target_label is e.g. "down"
>>> target_label = model.config.id2label[0]
>>> inputs["labels"] = torch.tensor([model.config.label2id[target_label]])
>>> loss = model(**inputs).loss( config )
Parameters
Wav2Vec2Bert Model with a frame classification head on top for tasks like Speaker Diarization.
Wav2Vec2Bert was proposed in wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations by Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, Michael Auli.
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 etc.).
This model is a PyTorch nn.Module sub-class. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
( input_features: Optional attention_mask: Optional = None labels: Optional = None output_attentions: Optional = None output_hidden_states: Optional = None return_dict: Optional = None ) → transformers.modeling_outputs.TokenClassifierOutput or tuple(torch.FloatTensor)
Parameters
torch.FloatTensor of shape (batch_size, sequence_length)) —
Float values of input raw speech waveform. Values can be obtained by loading a .flac or .wav audio file
into an array of type List[float] or a numpy.ndarray, e.g. via the soundfile library (pip install soundfile). To prepare the array into input_features, the AutoProcessor should be used for padding and
conversion into a tensor of type torch.FloatTensor. See Wav2Vec2BertProcessor.call() for details. torch.LongTensor of shape (batch_size, sequence_length), optional) —
Mask to avoid performing convolution and attention on padding token indices. Mask values selected in [0, 1]:
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 (batch_size,), optional) —
Labels for computing the sequence classification/regression loss. Indices should be in [0, ..., config.num_labels - 1]. If config.num_labels == 1 a regression loss is computed (Mean-Square loss), If
config.num_labels > 1 a classification loss is computed (Cross-Entropy). Returns
transformers.modeling_outputs.TokenClassifierOutput or tuple(torch.FloatTensor)
A transformers.modeling_outputs.TokenClassifierOutput 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 (Wav2Vec2BertConfig) and inputs.
loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Classification loss.
logits (torch.FloatTensor of shape (batch_size, sequence_length, config.num_labels)) — Classification scores (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.
The Wav2Vec2BertForAudioFrameClassification 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 AutoFeatureExtractor, Wav2Vec2BertForAudioFrameClassification
>>> from datasets import load_dataset
>>> import torch
>>> dataset = load_dataset("hf-internal-testing/librispeech_asr_demo", "clean", split="validation")
>>> dataset = dataset.sort("id")
>>> sampling_rate = dataset.features["audio"].sampling_rate
>>> feature_extractor = AutoFeatureExtractor.from_pretrained("facebook/w2v-bert-2.0")
>>> model = Wav2Vec2BertForAudioFrameClassification.from_pretrained("facebook/w2v-bert-2.0")
>>> # audio file is decoded on the fly
>>> inputs = feature_extractor(dataset[0]["audio"]["array"], return_tensors="pt", sampling_rate=sampling_rate)
>>> with torch.no_grad():
... logits = model(**inputs).logits
>>> probabilities = torch.sigmoid(logits[0])
>>> # labels is a one-hot array of shape (num_frames, num_speakers)
>>> labels = (probabilities > 0.5).long()( config )
Parameters
Wav2Vec2Bert Model with an XVector feature extraction head on top for tasks like Speaker Verification.
Wav2Vec2Bert was proposed in wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations by Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, Michael Auli.
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 etc.).
This model is a PyTorch nn.Module sub-class. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.
( input_features: Optional attention_mask: Optional = None output_attentions: Optional = None output_hidden_states: Optional = None return_dict: Optional = None labels: Optional = None ) → transformers.modeling_outputs.XVectorOutput or tuple(torch.FloatTensor)
Parameters
torch.FloatTensor of shape (batch_size, sequence_length)) —
Float values of input raw speech waveform. Values can be obtained by loading a .flac or .wav audio file
into an array of type List[float] or a numpy.ndarray, e.g. via the soundfile library (pip install soundfile). To prepare the array into input_features, the AutoProcessor should be used for padding and
conversion into a tensor of type torch.FloatTensor. See Wav2Vec2BertProcessor.call() for details. torch.LongTensor of shape (batch_size, sequence_length), optional) —
Mask to avoid performing convolution and attention on padding token indices. Mask values selected in [0, 1]:
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 (batch_size,), optional) —
Labels for computing the sequence classification/regression loss. Indices should be in [0, ..., config.num_labels - 1]. If config.num_labels == 1 a regression loss is computed (Mean-Square loss), If
config.num_labels > 1 a classification loss is computed (Cross-Entropy). Returns
transformers.modeling_outputs.XVectorOutput or tuple(torch.FloatTensor)
A transformers.modeling_outputs.XVectorOutput 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 (Wav2Vec2BertConfig) and inputs.
loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Classification loss.
logits (torch.FloatTensor of shape (batch_size, config.xvector_output_dim)) — Classification hidden states before AMSoftmax.
embeddings (torch.FloatTensor of shape (batch_size, config.xvector_output_dim)) — Utterance embeddings used for vector similarity-based retrieval.
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 + 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 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 Wav2Vec2BertForXVector 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 AutoFeatureExtractor, Wav2Vec2BertForXVector
>>> from datasets import load_dataset
>>> import torch
>>> dataset = load_dataset("hf-internal-testing/librispeech_asr_demo", "clean", split="validation")
>>> dataset = dataset.sort("id")
>>> sampling_rate = dataset.features["audio"].sampling_rate
>>> feature_extractor = AutoFeatureExtractor.from_pretrained("facebook/w2v-bert-2.0")
>>> model = Wav2Vec2BertForXVector.from_pretrained("facebook/w2v-bert-2.0")
>>> # audio file is decoded on the fly
>>> inputs = feature_extractor(
... [d["array"] for d in dataset[:2]["audio"]], sampling_rate=sampling_rate, return_tensors="pt", padding=True
... )
>>> with torch.no_grad():
... embeddings = model(**inputs).embeddings
>>> embeddings = torch.nn.functional.normalize(embeddings, dim=-1).cpu()
>>> # the resulting embeddings can be used for cosine similarity-based retrieval
>>> cosine_sim = torch.nn.CosineSimilarity(dim=-1)
>>> similarity = cosine_sim(embeddings[0], embeddings[1])
>>> threshold = 0.7 # the optimal threshold is dataset-dependent
>>> if similarity < threshold:
... print("Speakers are not the same!")