RM-Transformer:一个基于transformer的普通话语音识别模型

Xingmin Lu, Jianguo Hu, Shenhao Li, Yanyu Ding
{"title":"RM-Transformer:一个基于transformer的普通话语音识别模型","authors":"Xingmin Lu, Jianguo Hu, Shenhao Li, Yanyu Ding","doi":"10.1109/CCAI55564.2022.9807706","DOIUrl":null,"url":null,"abstract":"A network called RM-Transformer is proposed in this paper for Mandarin speech recognition. The proposed RMTransformer can make full use of features from different layers in the network instead of features solely from the top layer, which is used in the traditional models. Moreover, the proposed network has excellent capability in addressing the ambiguity problems caused by homophone phenomenon in Mandarin speech recognition task. Empirical evaluations have been conducted in two widely used datasets, which are Aishell-l and Aidatatang-200zh. Experimental results can verify the effectiveness of the proposed scheme.","PeriodicalId":340195,"journal":{"name":"2022 IEEE 2nd International Conference on Computer Communication and Artificial Intelligence (CCAI)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"RM-Transformer: A Transformer-based Model for Mandarin Speech Recognition\",\"authors\":\"Xingmin Lu, Jianguo Hu, Shenhao Li, Yanyu Ding\",\"doi\":\"10.1109/CCAI55564.2022.9807706\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"A network called RM-Transformer is proposed in this paper for Mandarin speech recognition. The proposed RMTransformer can make full use of features from different layers in the network instead of features solely from the top layer, which is used in the traditional models. Moreover, the proposed network has excellent capability in addressing the ambiguity problems caused by homophone phenomenon in Mandarin speech recognition task. Empirical evaluations have been conducted in two widely used datasets, which are Aishell-l and Aidatatang-200zh. Experimental results can verify the effectiveness of the proposed scheme.\",\"PeriodicalId\":340195,\"journal\":{\"name\":\"2022 IEEE 2nd International Conference on Computer Communication and Artificial Intelligence (CCAI)\",\"volume\":\"12 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-05-06\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 IEEE 2nd International Conference on Computer Communication and Artificial Intelligence (CCAI)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/CCAI55564.2022.9807706\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE 2nd International Conference on Computer Communication and Artificial Intelligence (CCAI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CCAI55564.2022.9807706","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

本文提出了一种用于汉语语音识别的RM-Transformer网络。本文提出的RMTransformer可以充分利用网络中不同层的特征,而不是像传统模型那样只使用来自顶层的特征。此外,该网络在解决汉语语音识别任务中同音现象引起的歧义问题方面具有出色的能力。本文在两个被广泛使用的数据集aishell - 1和Aidatatang-200zh上进行了实证评估。实验结果验证了该方案的有效性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
RM-Transformer: A Transformer-based Model for Mandarin Speech Recognition
A network called RM-Transformer is proposed in this paper for Mandarin speech recognition. The proposed RMTransformer can make full use of features from different layers in the network instead of features solely from the top layer, which is used in the traditional models. Moreover, the proposed network has excellent capability in addressing the ambiguity problems caused by homophone phenomenon in Mandarin speech recognition task. Empirical evaluations have been conducted in two widely used datasets, which are Aishell-l and Aidatatang-200zh. Experimental results can verify the effectiveness of the proposed scheme.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信