A novel stream-weight method for the multi-stream speech recognition system
A multi-stream speech recognition system is based on the combination of multiple complementary feature streams. Utilizing the fusion scheme of multi-stream, better performance was achieved in speech recognition system. The stream-weight method plays a very important role in the fusion collaborative scheme. The stream weights should be selected to be proportional to the feature stream reliability and informativeness. The posterior probability estimate is a measure of reliability,and the classification error is a measure of informativeness. The larger separation between class distributions in a given stream implies better discriminative power. The intra-class distances are an estimate of the class variance. The inter-and intra-class distances are combined to yield and estimate of the misclassification error for each stream. An unsupervised stream weight estimation method for multi-stream speech recognition system based on the computation of intra-and inter-class distances in each stream is proposed here. Experiments are conducted using Chinese Academy of Science speech database. Applying the new streamweigh algorithm, we achieve better fusion performance compared with some traditional fusion methods, and the word error rate was decreased by 6％.
multistream framework stream weighst interclass distance intra-class distance hidden markov models linear predictive cepstral coefficient mel-frequency cepstral coefficient
Hongyu Guo Xiaoqun Zhao Hongmiao Guo Hongyu Guo
School of Electronics and Information Engineering Tongji University Shanghai, China Engineering Training Center YanShan University Qinhuandao , Hebei Province, China School of Information Shanghai Ocean University Shanghai, China
国际会议
厦门
英文
179-182
2010-10-29（万方平台首次上网日期，不代表论文的发表时间）