Sixth International Conference on Spoken Language Processing (ICSLP 2000)

Beijing, China
October 16-20, 2000

Spoken Word Recognition Using the Artificial Evolution of a Set of Vocabulary

Tomio Takara, Eiji Nagaki

Department of Information Engineering, University of the Ryukyus, Senbaru, Nishihara, Okinawa, Japan

Hidden Markov models (HMMs) are widely used for automatic speech recognition. However, there is a problem still unresolved, i.e. how to design the optimal structure of the HMM. As an answer to this problem, we proposed the application of a genetic algorithm (GA) to search out such an optimal structure, and we showed this method to be effective for isolated word recognition. In these applications, the evolutions occurred at each word class independently. However, many isolated word recognition systems are performed using a set of vocabulary. Therefore, the artificial evolution using the vocabulary set is thought to be more e ective. In this paper, we propose the spoken word recognition using the artificial evolution of a set of vocabulary.


Full Paper

Bibliographic reference.  Takara, Tomio / Nagaki, Eiji (2000): "Spoken word recognition using the artificial evolution of a set of vocabulary", In ICSLP-2000, vol.1, 222-225.