Deterministic Annealing EM Algorithm in Acoustic Modeling for Speaker and Speech Recognition
Deterministic Annealing EM Algorithm in Acoustic Modeling for Speaker and Speech Recognition
复制标题
DOI：
10.1093/ietisy/e88-d.3.425
复制
发表时间：
2005-03
期刊：
影响因子：
--
通讯作者：
Yohei Itaya;H. Zen;Yoshihiko Nankaku;Chiyomi Miyajima;Keiichi Tokuda;T. Kitamura
中科院分区：
文献类型：
--
作者：
Yohei Itaya;H. Zen;Yoshihiko Nankaku;Chiyomi Miyajima;Keiichi Tokuda;T. Kitamura
研究了确定性退火法在说话人声学建模和语音识别中的有效性。虽然EM算法已被广泛用于逼近最大似然估计，但它存在初始化依赖的问题。为了缓解这一问题，提出了DAEM算法，并验证了其在人工小任务中的有效性。本文将DAEM算法应用于实际的语音识别任务：基于GMMS的说话人识别和基于HMMS的连续语音识别。实验结果表明，相对于标准EM算法和常规初始化算法，DAEM算法能够提高识别性能，特别是在连续语音识别的平坦开始训练中。
This paper investigates the effectiveness of the DAEM (Deterministic Annealing EM) algorithm in acoustic modeling for speaker and speech recognition. Although the EM algorithm has been widely used to approximate the ML estimates, it has the problem of initialization dependence. To relax this problem, the DAEM algorithm has been proposed and confirmed the effectiveness in artificial small tasks. In this paper, we applied the DAEM algorithm to practical speech recognition tasks: speaker recognition based on GMMs and continuous speech recognition based on HMMs. Experimental results show that the DAEM algorithm can improve the recognition performance as compared to the standard EM algorithm with conventional initialization algorithms, especially in the flat start training for continuous speech recognition.