A New Approach for Tone Recognition of Isolated Mandarin Syllables

Mandarin is tonal language. For Mandarin, tone identification is very important for speech recognition and pronunciation evaluation. This paper proposes a new approach for tone identification of isolated Mandarin syllables. The system employs a voiced/unvoiced (V/UV) classifier, an improved cepstral pitch detector and a powerful tone identification method. At the same time, the system takes several efficient measures to reduce the negative effect of creaky voice. Finally, the experimental results based on a very large database show that the tone recognition system classifies four tones very well