Abstract—Automatic speech recognition (ASR) is a technology that allows a computer to recognize the words spoken by a person through a telephone, microphone or other input devices. This paper focuses on recognizing numbers or digits because of their importance in banking system, customer queuing system, phone dialing system, and so on. We describe a speech recognition system for Myanmar digits based on the Hidden Markov Model (HMM) using HTK Tools. Our system recognizes the speech utterance by converting the speech waveform into a set of feature vectors using Mel Frequency Cepstral Coefficients (MFCC) technique. We use HMM-based acoustic and language models. Experiment is carried out to evaluate the performance of our speech recognizer with both context independent and context dependent models and yields significant results.
Index Terms—Hidden Markov model, MFCC, Myanmar digits, speech recognition.
The authors are with the Department of Information, Computer, and Communication Technology, Sirindhorn International Institute of Technology, Thammasat University, Thailand (e-mail: zin2tun@gmail.com, gun@siit.tu.ac.th).
[PDF]
Cite:Zin Zin Tun and Gun Srijuntongsiri, "A Speech Recognition System for Myanmar Digits," International Journal of Information and Electronics Engineering vol. 6, no. 3, pp. 210-213, 2016.