关键词: 隐马尔可夫模型, 遗传算法, 语音识别, 声学模型拓扑结构, 贝叶斯信息准则

Abstract: Aiming at the current situation of selecting acoustic model topologies empirically or heuristically, an acoustic model topology optimization algorithm based on standard Genetic Algorithm（GA） is proposed. Compared with the previous similar algorithms, it is an automatic one with the characteristics of optimizing model state number and kernel numbers simultaneously and rejecting uniform allocation of Gaussian kernels. Experiments using Bayesian Information Criterion（BIC） as the objective function on TIDigits corpus show that, the speech recognition system whose model topologies are optimized using the GA based algorithm is able to obtain higher recognition performance with smaller number of parameters compared with the baselines built in the conventional way, which indicates that the GA based algorithm is an effective tool of optimizing acoustic model topologies.

Key words: Hidden Markov Model（HMM）, Genetic Algorithm（GA）, speech recognition, acoustic model topology, Bayesian Information Criterion（BIC）