.png){kind=icon}
返回
Mutual-learning sequence-level knowledge distillation for automatic speech recognition
.png){kind=icon}
OA
AI
DOI：10.1016/j.neucom.2020.11.025.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Automatic speech recognition (ASR) is a crucial technology for man-machine interaction. End-to-end models have been studied recently in deep learning for ASR. However, these models are not suitable for the practical application of ASR due to their large model sizes and computation costs. To address this issue, we propose a novel mutual-learning sequence-level knowledge distillation framework enjoying distinct student structures for ASR. Trained mutually and simultaneously, each student learns not only from the pre-trained teacher but also from its distinct peers, which can improve the generalization capability of the whole network, through making up for the insufficiency of each student and bridging the gap between each student and the teacher. Extensive experiments on the TIMIT and large LibriSpeech corpuses show that, compared with the state-of-the-art methods, the proposed method achieves an excellent balance between recognition accuracy and model compression. (C) 2020 Elsevier B.V. All rights reserved.
Keyword:
Automatic speech recognition (ASR)
Model compression
Knowledge distillation (KD)
Mutual learning
Connectionist temporal classification (CTC)
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
6.5
论文数:
2.5W
被引数:
6.5W
机构
引用论文
Allele-Specific Transcriptional Regulation of Shoot Regeneration in Hybrid Poplar杂种杨树中再生芽的等位基因特异性转录调控.png)
Forests
OAAI
ForestsIF0
OAAI分享收藏Geosites in Karamay city, Xinjiang Uygur Autonomous Region, northwest China新疆维吾尔自治区西北部中国克拉玛依市的地质遗迹Geoheritage
GeoheritageIF0
PREAI
分享收藏Compressing CNN-DBLSTM models for OCR with teacher-student learning and Tucker decompositionPATTERN RECOGNITION
PATTERN RECOGNITIONIF7.6
PREAI
分享收藏
查看更多内容