Self-Attention Transducers for End-to-End Speech Recognition
arXiv:1909.13037 · doi:10.21437/Interspeech.2019-2203
Abstract
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.
References in corpus (2)
Cited by in corpus (11)
- Exploring Transformers for Large-Scale Speech Recognition
- Spike-Triggered Non-Autoregressive Transformer for End-to-End Speech Recognition
- SAN-M: Memory Equipped Self-Attention for End-to-End Speech Recognition
- Attention-based Transducer for Online Speech Recognition
- Synchronous Transformers for End-to-End Speech Recognition
- A Transformer with Interleaved Self-attention and Convolution for Hybrid Acoustic Models
- Transformer with Bidirectional Decoder for Speech Recognition
- Rnn-transducer with language bias for end-to-end Mandarin-English code-switching speech recognition
- Self-and-Mixed Attention Decoder with Deep Acoustic Structure for Transformer-based LVCSR
- Conv-Transformer Transducer: Low Latency, Low Frame Rate, Streamable End-to-End Speech Recognition
- Exploration of Audio Quality Assessment and Anomaly Localisation Using Attention Models