SEQ-former: A context-enhanced and efficient automatic speech recognition framework

Qinglin Meng, Min Liu, Kaixun Huang, Kun Wei, Lei Xie, Zongfeng Quan, Weihong Deng, Quan Lu, Ning Jiang, Guoqing Zhao

Research output: Contribution to journalConference articlepeer-review

Abstract

Contextual information is crucial for automatic speech recognition (ASR). Effective utilization of contextual information can improve the accuracy of ASR systems. To improve the model's ability to capture this information, we propose a novel ASR framework called SEQ-former, emphasizing simplicity, efficiency, and quickness. We incorporate a Prediction Decoder Network and a Shared Prediction Decoder Network to enhance contextual capabilities. To further increase efficiency, we use intermediate CTC and CTC Spike Reduce Methods to guide attention masks and reduce redundant peaks. Our approach demonstrates state-of-the-art performance on the AiShell-1 dataset, improves decoding efficiency, and delivers competitive results on LibriSpeech. Additionally, it optimizes 6.3% over 11,000 hours of private data compared to Efficient Conformer.

Original languageEnglish
Pages (from-to)212-216
Number of pages5
JournalProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
DOIs
StatePublished - 2024
Event25th Interspeech Conferece 2024 - Kos Island, Greece
Duration: 1 Sep 20245 Sep 2024

Keywords

  • Blank-regularized CTC
  • Prediction Decoder Network
  • SEQ-former
  • contextual information
  • speech recognition

Fingerprint

Dive into the research topics of 'SEQ-former: A context-enhanced and efficient automatic speech recognition framework'. Together they form a unique fingerprint.

Cite this