Supervised Attention in Sequence-to-Sequence Models for Speech Recognition
This addresses a key bottleneck in speech recognition models for improving accuracy, though it is incremental as it builds on existing attention mechanisms.
The paper tackles the problem of attention weights in sequence-to-sequence speech recognition models not aligning well with actual acoustic-token alignments, and it introduces a supervised attention loss to improve this, resulting in significant performance gains.
Attention mechanism in sequence-to-sequence models is designed to model the alignments between acoustic features and output tokens in speech recognition. However, attention weights produced by models trained end to end do not always correspond well with actual alignments, and several studies have further argued that attention weights might not even correspond well with the relevance attribution of frames. Regardless, visual similarity between attention weights and alignments is widely used during training as an indicator of the models quality. In this paper, we treat the correspondence between attention weights and alignments as a learning problem by imposing a supervised attention loss. Experiments have shown significant improved performance, suggesting that learning the alignments well during training critically determines the performance of sequence-to-sequence models.