计算机工程2011，Vol.37Issue(22)：268-269,272,3.
基于发音特征的音视频融合语音识别模型
Audio Visual Fusion Speech Recognition Model Based on Articulatory Feature
摘要
Abstract
A multi-stream Dynamic Bayesian Nctwork(DBN) model(AF_AV_DBN) based on Articulatory Featurc(AF) is proposed for audio visual speech recognition. Conditional probability distribution or each node and the degrcc of asynchrony between the AFs are defined, and speech recognition experiments arc carried out on an audio visual connected digit database. Compared with ihc audio-only AF_A_DBN model, the state synchronous DBN model and the state asynchronous DBN model, the designed AF^AV_DBN model gets the highest recognition rate under various signal to noise ratios, and is more robust to background noise.关键词
:动态贝叶斯网络/发音特征/音视频融合/语音识别/异步Key words
Dynamic Bayesian Network(DBN)/ articulatory feature/ audio visual fusion/ speech recognition/ asynchronous分类
信息技术与安全科学引用本文复制引用
吴鹏,蒋冬梅,王风娜,Hichem SAHLI,Werner VERHEIST..基于发音特征的音视频融合语音识别模型[J].计算机工程,2011,37(22):268-269,272,3.基金项目
国家自然科学基金资助项目(60703104) （60703104）
陕西省自然科学基金资助项目(SJ08F28) （SJ08F28）
西北工业大学基础研究基金资助项目(JC200943) （JC200943）
