torchaudio.functional.rnnt_loss

torchaudio.functional.rnnt_loss(logits: Tensor, targets: Tensor, logit_lengths: Tensor, target_lengths: Tensor, blank: int = -1, clamp: float = -1, reduction: str = 'mean')

Compute the RNN Transducer loss from Sequence Transduction with Recurrent Neural Networks [Graves, 2012].

The RNN Transducer loss extends the CTC loss by defining a distribution over output sequences of all lengths, and by jointly modelling both input-output and output-output dependencies.

Parameters:

• logits (Tensor) – Tensor of dimension (batch, max seq length, max target length + 1, class) containing output from joiner

• targets (Tensor) – Tensor of dimension (batch, max target length) containing targets with zero padded

• logit_lengths (Tensor) – Tensor of dimension (batch) containing lengths of each sequence from encoder

• target_lengths (Tensor) – Tensor of dimension (batch) containing lengths of targets for each sequence

• blank (int, optional) – blank label (Default: -1)

• clamp (float, optional) – clamp for gradients (Default: -1)

• reduction (string, optional) – Specifies the reduction to apply to the output: "none" | "mean" | "sum". (Default: "mean")

Returns:

Loss with the reduction option applied. If reduction is "none", then size (batch), otherwise scalar.

Return type:

Tensor