基于无状态预测网络的rnn换能器

M. Ghodsi, Xiaofeng Liu, J. Apfel, Rodrigo Cabrera, Eugene Weinstein
{"title":"基于无状态预测网络的rnn换能器","authors":"M. Ghodsi, Xiaofeng Liu, J. Apfel, Rodrigo Cabrera, Eugene Weinstein","doi":"10.1109/ICASSP40776.2020.9054419","DOIUrl":null,"url":null,"abstract":"The RNN-Transducer (RNNT) outperforms classic Automatic Speech Recognition (ASR) systems when a large amount of supervised training data is available. For low-resource languages, the RNNT models overfit, and can not directly take advantage of additional large text corpora as in classic ASR systems.We focus on the prediction network of the RNNT, since it is believed to be analogous to the Language Model (LM) in the classic ASR systems. We pre-train the prediction network with text-only data, which is not helpful. Moreover, removing the recurrent layers from the prediction network, which makes the prediction network stateless, performs virtually as well as the original RNNT model, when using wordpieces. The stateless prediction network does not depend on the previous output symbols, except the last one. Therefore it simplifies the RNNT architectures and the inference.Our results suggest that the RNNT prediction network does not function as the LM in classical ASR. Instead, it merely helps the model align to the input audio, while the RNNT encoder and joint networks capture both the acoustic and the linguistic information.","PeriodicalId":13127,"journal":{"name":"ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","volume":"39 1","pages":"7049-7053"},"PeriodicalIF":0.0000,"publicationDate":"2020-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"77","resultStr":"{\"title\":\"Rnn-Transducer with Stateless Prediction Network\",\"authors\":\"M. Ghodsi, Xiaofeng Liu, J. Apfel, Rodrigo Cabrera, Eugene Weinstein\",\"doi\":\"10.1109/ICASSP40776.2020.9054419\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The RNN-Transducer (RNNT) outperforms classic Automatic Speech Recognition (ASR) systems when a large amount of supervised training data is available. For low-resource languages, the RNNT models overfit, and can not directly take advantage of additional large text corpora as in classic ASR systems.We focus on the prediction network of the RNNT, since it is believed to be analogous to the Language Model (LM) in the classic ASR systems. We pre-train the prediction network with text-only data, which is not helpful. Moreover, removing the recurrent layers from the prediction network, which makes the prediction network stateless, performs virtually as well as the original RNNT model, when using wordpieces. The stateless prediction network does not depend on the previous output symbols, except the last one. Therefore it simplifies the RNNT architectures and the inference.Our results suggest that the RNNT prediction network does not function as the LM in classical ASR. Instead, it merely helps the model align to the input audio, while the RNNT encoder and joint networks capture both the acoustic and the linguistic information.\",\"PeriodicalId\":13127,\"journal\":{\"name\":\"ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)\",\"volume\":\"39 1\",\"pages\":\"7049-7053\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-05-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"77\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICASSP40776.2020.9054419\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICASSP40776.2020.9054419","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 77

摘要

当有大量的监督训练数据可用时,rnn -换能器(RNNT)优于经典的自动语音识别(ASR)系统。对于低资源语言,RNNT模型过拟合,并且不能像经典ASR系统那样直接利用额外的大型文本语料库。我们将重点放在RNNT的预测网络上,因为它被认为类似于经典ASR系统中的语言模型(LM)。我们用纯文本数据预训练预测网络,这是没有帮助的。此外,从预测网络中去除循环层,使预测网络无状态,在使用词块时,几乎与原始RNNT模型一样好。无状态预测网络不依赖于之前的输出符号,除了最后一个。因此,它简化了RNNT体系结构和推理。我们的研究结果表明,RNNT预测网络并不像经典ASR中的LM那样起作用。相反,它只是帮助模型与输入音频对齐,而RNNT编码器和联合网络同时捕获声学和语言信息。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Rnn-Transducer with Stateless Prediction Network
The RNN-Transducer (RNNT) outperforms classic Automatic Speech Recognition (ASR) systems when a large amount of supervised training data is available. For low-resource languages, the RNNT models overfit, and can not directly take advantage of additional large text corpora as in classic ASR systems.We focus on the prediction network of the RNNT, since it is believed to be analogous to the Language Model (LM) in the classic ASR systems. We pre-train the prediction network with text-only data, which is not helpful. Moreover, removing the recurrent layers from the prediction network, which makes the prediction network stateless, performs virtually as well as the original RNNT model, when using wordpieces. The stateless prediction network does not depend on the previous output symbols, except the last one. Therefore it simplifies the RNNT architectures and the inference.Our results suggest that the RNNT prediction network does not function as the LM in classical ASR. Instead, it merely helps the model align to the input audio, while the RNNT encoder and joint networks capture both the acoustic and the linguistic information.
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:481959085
Book学术官方微信