Year: 2022
Type: preprint
Abstract: 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.
Source: arXiv (Cornell University)
Authors: Yang, Gene-Ping, Tang, Hao
Institution: University of Edinburgh
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4303941907