Implementation:Speechbrain Speechbrain Train AISHELL1 Seq2Seq

Knowledge Sources	SpeechBrain
Domains	ASR, Training
Last Updated	2026-02-09 00:00 GMT

Overview

Concrete tool for training a sequence-to-sequence ASR model on the AISHELL-1 dataset provided by the SpeechBrain library.

Description

This recipe defines the ASR class (subclass of sb.Brain) for attention-based sequence-to-sequence speech recognition on Mandarin Chinese. The encoder processes waveform features through a CRDNN, while the decoder uses an embedding layer and RNN with attention. The model is jointly trained with CTC and NLL (negative log-likelihood) losses, with CTC used only during early epochs. Beam search decoding is applied at validation and test stages.

Usage

Use this recipe to train a seq2seq ASR model with CTC+attention on the AISHELL-1 Mandarin Chinese dataset. Requires the corresponding hyperparameter YAML file and data preparation script.

Code Reference

Source Location

• Repository: SpeechBrain
• File: recipes/AISHELL-1/ASR/seq2seq/train.py

Signature

class ASR(sb.Brain):
    def compute_forward(self, batch, stage):
        """Forward computations from the waveform batches to the output probabilities."""
        ...
    def compute_objectives(self, predictions, batch, stage):
        """Computes the loss (CTC+NLL) given predictions and targets."""
        ...

Import

# Run as recipe script
python recipes/AISHELL-1/ASR/seq2seq/train.py hparams/train.yaml --data_folder /path/to/aishell

I/O Contract

Inputs

Name	Type	Required	Description
batch.sig	torch.Tensor	Yes	Input waveform signal
batch.tokens_bos	torch.Tensor	Yes	Target token sequence with BOS prefix
batch.tokens_eos	torch.Tensor	Yes	Target token sequence with EOS suffix
batch.tokens	torch.Tensor	Yes	Target token sequence (for CTC)

Outputs

Name	Type	Description
p_ctc	torch.Tensor	CTC log-probabilities (during early training epochs)
p_seq	torch.Tensor	Seq2seq log-probabilities from attention decoder
wav_lens	torch.Tensor	Relative waveform lengths
p_tokens	list	Beam search hypotheses (at validation/test)

Usage Examples

python train.py hparams/train.yaml --data_folder /path/to/AISHELL-1

Related Pages

• Principle:Speechbrain_Speechbrain_Seq2Seq_ASR_Training