State-of-the-Art Speech Recognition Using Multi-Stream Self-Attention With Dilated 1D Convolutions
This work addresses a key bottleneck in speech recognition for applications requiring high accuracy, though it appears incremental as it builds on existing self-attention methods.
The paper tackled the challenge of applying self-attention to speech recognition by proposing a multi-stream self-attention architecture with dilated 1D convolutions, achieving a state-of-the-art word error rate of 2.2% on the LibriSpeech test-clean dataset.
Self-attention has been a huge success for many downstream tasks in NLP, which led to exploration of applying self-attention to speech problems as well. The efficacy of self-attention in speech applications, however, seems not fully blown yet since it is challenging to handle highly correlated speech frames in the context of self-attention. In this paper we propose a new neural network model architecture, namely multi-stream self-attention, to address the issue thus make the self-attention mechanism more effective for speech recognition. The proposed model architecture consists of parallel streams of self-attention encoders, and each stream has layers of 1D convolutions with dilated kernels whose dilation rates are unique given stream, followed by a self-attention layer. The self-attention mechanism in each stream pays attention to only one resolution of input speech frames and the attentive computation can be more efficient. In a later stage, outputs from all the streams are concatenated then linearly projected to the final embedding. By stacking the proposed multi-stream self-attention encoder blocks and rescoring the resultant lattices with neural network language models, we achieve the word error rate of 2.2% on the test-clean dataset of the LibriSpeech corpus, the best number reported thus far on the dataset.