Large-Vocabulary Continuous Speech Recognition of Lhasa Tibetan

Author:

Li Guan Yu1,Yu Hong Zhi1

Affiliation:

1. Northwest University for Nationalities

Abstract

The framework of auto speech recognition of Lhasa dialect was established in this paper. Phoneme was chosen as the basic unit for modeling. Then, phonemes set of Lhasa dialect and their Latin transliteration were designed. There were 5568 frequently used monosyllables in the vocabulary. Hidden Markov Models of triphones were established and trained by use of HTK. Word error rate (WER) was 21.81% under the optimal situation.

Publisher

Trans Tech Publications, Ltd.

Reference12 articles.

1. Gongque Jiangcuo, Theory of Tibetan, Tibetan Studies . 1997 III.

2. Kelsang Jumian, Practical Tibetan grammar course. Sichuan minorities press. November 2004 edition.

3. Han Qinghua, Yu Hongzhi. Anduo Tibetan speaker-independent isolated words speech recognition based on HMMs. Software Guide . 2010 07.

4. Pei Chun Bao. The Tibetan language speech recognition technology based on the standard Lhasa, Master Thesis in Tibet University , (2009).

5. The HTK Book(for HTK Version 3. 4). Cambridge University Engineering Department. (2009).

Cited by 2 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3