期刊文献+

语音关键词检测中置信测度方法研究综述

An Overview of Confidence Measure for Spoken Term Detection
下载PDF
导出
摘要 语音关键词检测是指在语音文档中寻找并定位特定的词的技术,输入所需查询的关键词通常是以文本的形式给出。作为语音文档分析等技术的核心部分,语音关键词检测始终是语音处理领域研究的热点。置信测度计算是关键词检测技术的重要组成部分,对确认正确检出的关键词及拒绝误识都起到决定性作用,置信测度的优劣对检测系统性能有着直接的影响。介绍并总结了语音关键词检测中测置信测度方法的研究工作,提供了详尽的参考文献。 Spoken term detection( STD) is the task which aims to locate all occurrences of terms queried by a user in large audio archives,and the terms are usually in the format of text. STD is one of the key components for spoken document analysis,and STD is always a focus in the field of speech processing. Confidence measure is a crucial part of STD,and it plays an important role in rejecting false alarms. The merit of confidence measure can affect the performance of STD directly. This paper introduces and summarizes the research work of confidence measure for STD,and provides with plenty of references.
出处 《智能计算机与应用》 2014年第2期10-15,共6页 Intelligent Computer and Applications
基金 国家重点基础研究发展计划(973)(2007CB311104) 国家自然科学基金重大研究计划项目(91120303)
关键词 检测 置信测度 词表外词 Spoken Term Detection Confidence Measure Out-of-vocabulary Term
  • 相关文献

参考文献1

二级参考文献16

  • 1孟莎,余鹏,Frank Seide,刘加.基于后验概率词格的汉语自然对话语音索引[J].清华大学学报(自然科学版),2008,48(S1):673-677. 被引量:2
  • 2周梁,高鹏,丁鹏,徐波.语音识别准确率与检索性能的关联性研究[J].中文信息学报,2006,20(3):99-104. 被引量:2
  • 3M. Saraclar and R. Sproat. Lattice-based Search for Spoken Utterance[C]//Proceeding of Human Language Technology Conference. Boston, 2004: 129-136. 被引量:1
  • 4C. Chelba and A. Acero. Position specific posterior lattices for indexing speech [C]//Proceedings of the 43rd Annual Meeting on Association for Computational Linguistics. Ann Arbor, 2005: 443-450. 被引量:1
  • 5F. Seide, P. Yu and Y. Shi. Towards Spoken-Document Retrieval for the Enterprise: Approximate Word- Lattice Indexing with Text Indexers [C]//Proceeding of IEEE Workshop on Automatic Speech Recognition and Understanding. Kyoto, 2007: 629-634. 被引量:1
  • 6B. Logan, P. Moreno, J. M. Van Tong et al. An Experimental Study of an Audio Indexing System for the Web [C]//Proceeding of Sixth International Conference on Spoken Language Processing. Beijing, 2000: 676-679. 被引量:1
  • 7K. Ng. Subword-Based Approaches for Spoken Document Retrieval [D]. Ph. D. thesis, Massachusetts In- stitute of Technology, 2000. 被引量:1
  • 8P. Yu and F. Seide. A Hybrid Word/Phoneme-based Approach for Improved Vocabulary-independent Search in Spontaneous Speech [C]//Proceeding of Sixth International Conference on Spoken Language Processing, Korean, 2004: 293-296. 被引量:1
  • 9J. Shao, P Yu, Q. Zhao, Y. Yan. F. Seide. Towards Vocabulary-Independent Speech Indexing for Large-Scale Repositories [C]//Proceeding of Inter- speech. Brisbane, 2008:2150-2153. 被引量:1
  • 10H. M. Wang, H. Meng, P. Schone, B. Chen, W. K. Lo. Multi-Scale Audio Indexing for Translingual Spoken Document Retrieval [C]//Proceedings of IEEE Interna- tional Conference on Acoustics, Speech and Signal Processing. Salt Lake City, 2001: 605-608. 被引量:1

共引文献7

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