期刊文献+
任意字段
题名或关键词
题名
关键词
文摘
作者
第一作者
机构
刊名
分类号
参考文献
作者简介
基金资助
栏目信息

汉语语音识别中语言模型的并行优化 被引量：1

Parallel Optimization Method in Language Model for Mandarin Speech Recognition
下载PDF
导出
摘要 研究了汉语语音识别中语言模型的并行优化问题,在已有汉语语音识别有调拼音模型和多核体系结构的基础上,对其算法进行了并行优化,并优化了查询机制和前后向算法。试验表明,该方法取得了很好的效果,识别速度有了很大改进。 In order to improve performance and throughput of speech recognition system, parallel optimization method in language model is proposed based on mandarin N-gram model and multi-core systems. The searching algorithm is optimized, and the forward-algorithm and backward-algorithm are parallelized on different CPU-cores. The ideal properties of the optimized speech recognition system provide users with high speed and throughput. It makes real-time recognition much easier.
作者 金玮 孙甲松
机构地区 清华大学电子工程系
出处 《电声技术》 2010年第8期49-52,共4页 Audio Engineering
关键词 关键词语音识别 N元有调拼音文法模型 并行优化 speech recognition N-gram model for Pinyin with tone parallel optimization
分类号 TN912 [电子电信—通信与信息系统]
  • 相关文献

参考文献6

  • 1孙甲松,王作英,吴及.汉语语音识别中的有调拼音多候选问题[C]//第六届全国人际语音通讯学术会议论文集.深圳:中国计算机学会人工智能与模式识别专业委员会等6专业委员会,2001:243-247.
  • 2王作英,肖熙.基于段长分布的HMM语音识别模型[J].电子学报,2004,32(1):46-49. 被引量：42
  • 3林建臻,孙甲松,王作英.汉语语音识别的抗噪性前端算法及性能分析[J].电声技术,2004,28(3):45-48. 被引量：1
  • 4JELINEK F.Self-organized language modeling for speech recognition[M]. San Mateo: Readings in Speech Recognition, Morgan-Kaufmann, 1990 : 450-506.
  • 5BELLMAN R E. Dynamic programming[M].Princeton : Princeton University Press, 1957.
  • 6GUODONG Z, KIMTENG L. Interpolation of n-gram and mutual-information based trigger pair language models for Mandarin speech recognition[J]. Computer Speech and Language, 1999,13(2) : 125-141.

二级参考文献13

  • 1齐士钤 张家禄.汉语普通话辅音音长分析[J].声学学报,1982,(1):8-13.
  • 2王作英.基于段长分布的HMM语音识别模型 [A]..第二届全国汉字汉语识别会议 [C].庐山,1989.9.
  • 3Satoru Tsuge, Toshiaki Fukada, Harald Singer. Speaker Normalized Spectral Subband Parameters For Noise Robust Recognition. ICASSP 1999.
  • 4Steven F. Boll. Suppression of Acoustic Noise in Speech Using Spectral Subtraction. IEEE TRANSACTIONS ON SPEECH AND AUDIO PROCESSING, 1979.
  • 5Agarwal A., Cheng Y.M.. Two-stage Mel-Warped Wiener Filter for Robust Speech Recognition. Proc.ASRU'99,1999.
  • 6Hynek Hermansky. Exploring Temporal Domain for Robustness in Speech Recognition. Invited paper, the 15th International Congress on Acoustics, 1995.
  • 7Agarwal A., Cheng Y. M.. Two-stage Mel-Warped Wiener Filter for Robust Speech Recognition. Proceeding of IEEE workshop on Automatic Speech Recognition and Understanding, Keystone, 1999.
  • 8Li Deng, Jasha Droppo, Alex Acero. A Bayesian Approach to Speech Feature Enhancement Using the Dynamic Cepstral Prior. ICASSP 2002.
  • 9Qi Li,Frank K.Soong, Olivier Siohan. An Auditory system-based feature for robust speech recognition. Eurospeech 2001.
  • 10Xiaodong Cui, Markus Iseli, Qifeng Zhu,Abeer Alwan. Evaluation of Noise Robust Features on the Aurora Databases. ICSLP 2002.

共引文献41

同被引文献4

引证文献1

二级引证文献1

电声技术
2010年 第8期

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部