2021 IEEE International Conference on Acoustics, Speech and Signal Processing
6-11 June 2021 • Toronto, Ontario, Canada
Extracting Knowledge from Information
| Paper ID | SPE-10.5 | ||
| Paper Title | HEAD-SYNCHRONOUS DECODING FOR TRANSFORMER-BASED STREAMING ASR | ||
| Authors | Mohan Li, Cătălin Zorilă, Rama Doddipatla, Toshiba Cambridge Research Laboratory, United Kingdom | ||
| Session | SPE-10: Speech Recognition 4: Transformer Models 2 | ||
| Location | Gather.Town | ||
| Session Time: | Tuesday, 08 June, 16:30 - 17:15 | ||
| Presentation Time: | Tuesday, 08 June, 16:30 - 17:15 | ||
| Presentation | Poster | ||
| Topic | Speech Processing: [SPE-LVCR] Large Vocabulary Continuous Recognition/Search | ||
| IEEE Xplore Open Preview | Click here to view in IEEE Xplore | ||
| Abstract | Online Transformer-based automatic speech recognition (ASR) systems have been extensively studied due to the increasing demand for streaming applications. Recently proposed Decoder-end Adaptive Computation Steps (DACS) algorithm for online Transformer ASR was shown to achieve state-of-the-art performance and outperform other existing methods. However, like any other online approach, the DACS-based attention heads in each of the Transformer decoder layers operate independently (or asynchronously) and lead to diverged attending positions. Since DACS employs a truncation threshold to determine the halting position, some of the attention weights are cut off untimely and might impact the stability and precision of decoding. To overcome these issues, here we propose a head-synchronous (HS) version of the DACS algorithm, where the boundary of attention is jointly detected by all the DACS heads in each decoder layer. ASR experiments on Wall Street Journal (WSJ), AIShell-1 and Librispeech show that the proposed method consistently outperforms vanilla DACS and achieves state-of-the-art performance. We will also demonstrate that HS-DACS has reduced decoding cost when compared to vanilla DACS. | ||