Language Modelling Approaches for Turkish Large Vocabulary Continuous Speech Recognition Based on Lattice Rescoring

E. Arisoy, M. Saraçlar
{"title":"Language Modelling Approaches for Turkish Large Vocabulary Continuous Speech Recognition Based on Lattice Rescoring","authors":"E. Arisoy, M. Saraçlar","doi":"10.1109/SIU.2006.1659773","DOIUrl":null,"url":null,"abstract":"In this paper, we have tried some language modelling approaches for large vocabulary continuous speech recognition (LVCSR) of Turkish. The agglutinative nature of Turkish makes Turkish a challenging language in terms of speech recognition since it is impossible to include all possible words in the recognition lexicon. Therefore, instead of using words as recognition units, we use a data-driven sub-word approach called morphs. This method was previously applied to Finnish, Estonian and Turkish and promising recognition results were achieved compared to words as recognition units. In our database, we obtained word error rates (WER) of 38.8% for the baseline word-based model and 33.9% for the baseline morph-based model. In addition, we tried some new methods. Recognition lattice outputs of each model were rescored with the root-based and root-class-based models for the word-based case and first-morph-based model for the morph-based case. The word-root composition approach achieves a 0.5% increase in the recognition performance. However, other two approaches fail due to the non-robust estimates over the baseline models","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"81 4 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2006 IEEE 14th Signal Processing and Communications Applications","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SIU.2006.1659773","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2

Abstract

In this paper, we have tried some language modelling approaches for large vocabulary continuous speech recognition (LVCSR) of Turkish. The agglutinative nature of Turkish makes Turkish a challenging language in terms of speech recognition since it is impossible to include all possible words in the recognition lexicon. Therefore, instead of using words as recognition units, we use a data-driven sub-word approach called morphs. This method was previously applied to Finnish, Estonian and Turkish and promising recognition results were achieved compared to words as recognition units. In our database, we obtained word error rates (WER) of 38.8% for the baseline word-based model and 33.9% for the baseline morph-based model. In addition, we tried some new methods. Recognition lattice outputs of each model were rescored with the root-based and root-class-based models for the word-based case and first-morph-based model for the morph-based case. The word-root composition approach achieves a 0.5% increase in the recognition performance. However, other two approaches fail due to the non-robust estimates over the baseline models
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
基于点阵评分的土耳其语大词汇连续语音识别语言建模方法
本文针对土耳其语的大词汇量连续语音识别(LVCSR),尝试了几种语言建模方法。土耳其语的黏着性使得土耳其语在语音识别方面成为一种具有挑战性的语言,因为在识别词典中不可能包括所有可能的单词。因此,我们不使用词作为识别单位,而是使用数据驱动的子词方法,称为morphs。该方法之前应用于芬兰语、爱沙尼亚语和土耳其语,与以单词为识别单位相比,取得了令人满意的识别结果。在我们的数据库中,我们获得了基线基于词的模型的单词错误率(WER)为38.8%,基线基于词形的模型为33.9%。此外,我们还尝试了一些新的方法。每个模型的识别格输出分别使用基于根和基于根类的模型对基于词的情况进行重建,使用基于第一形态的模型对基于形态的情况进行重建。词根合成方法的识别性能提高了0.5%。然而,另外两种方法由于对基线模型的非稳健估计而失败
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Peer-to-Peer Multipoint Video Conferencing Using Layered Video Determination of Product Surface Quality Watermarking Tools for Turkish Texts By Using Darlington Topology Improvement of In-Band Gain for the Log Domain Filters Dual Wideband Antenna Analysis for Linear FMCW Radar Applications
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1