Tone Recognition of Continuous Mandarin Speech Based on Binary-Class SVMs

Tone is an essential component for word formation in all tone languages. It plays a very important role in the transmission of information in speech communication. In this paper, we look at using support vector machines (SVMs) for automatic tone recognition in continuously spoken Mandarin. Wavelet transform and Teager Energy Operation (TEO) are used to detect the voiced segments. Considerable improvement has been achieved by adopting binary-SVMs scheme in a speaker-independent Mandarin tone recognition system.