Self-Attention Transducers for End-to-End Speech Recognition
This work addresses the need for more efficient and parallelizable speech recognition systems, particularly for Mandarin Chinese, though it is incremental as it builds on existing transducer frameworks.
The authors tackled the problem of parallelization difficulty in recurrent neural network transducers for end-to-end speech recognition by proposing a self-attention transducer, achieving a 21.3% relative reduction in character error rate on the AISHELL-1 Mandarin dataset.
Recurrent neural network transducers (RNN-T) have been successfully applied in end-to-end speech recognition. However, the recurrent structure makes it difficult for parallelization . In this paper, we propose a self-attention transducer (SA-T) for speech recognition. RNNs are replaced with self-attention blocks, which are powerful to model long-term dependencies inside sequences and able to be efficiently parallelized. Furthermore, a path-aware regularization is proposed to assist SA-T to learn alignments and improve the performance. Additionally, a chunk-flow mechanism is utilized to achieve online decoding. All experiments are conducted on a Mandarin Chinese dataset AISHELL-1. The results demonstrate that our proposed approach achieves a 21.3% relative reduction in character error rate compared with the baseline RNN-T. In addition, the SA-T with chunk-flow mechanism can perform online decoding with only a little degradation of the performance.