基于语音识别与特征的无监督语音模式提取-计算机工程.pdfVIP

40 5 2014 5 Vol.40 No.5 Computer Engineering May 2014 ·· 2014 A TN912.34 ( 100190) 2 2 Unsupervised Speech Pattern Extraction Based on Speech Recognition and Feature ZHANG Zhen, ZHAO Qing-wei, YAN Yong-hong (Key Laboratory of Speech Acoustics and Content Understanding, Chinese Academy of Sciences, Beijing 100190, China) AbstractThis paper proposes the unsupervised method based on both speech recognition system and feature-based system to search for the speech patterns. In speech recognition system, the alternative results of the speech recognition system decoder are used to search audio patterns with segmental dynamic time warping algorithm. Then graph clustering algorithm is used, as well as confidence estimation algorithm, to improve the performance of the system. It also proposes the system based on feature only without any knowledge resource. In the final, the performances of the two systems on both radio and television news and spoken dialogue sets are compared. The speech recognition system achieves better performance, and the feature based system can be used on many languages. Key wordsspeech recognition; speech pattern discovery; segmental dynamic time warping algorithm; graph clustering algorithm; phoneme loop calculation of posterior probability DOI: 10.3969/j.issn2014.05.054