Authors : Kasiprasad Mannepalli, P. Narahari Sastry and V. Rajesh
Abstract: Speech recognition systems are used in many applications, it is crucial for the speech recognition systems to be able to deal with accented speakers. The speech recognition systems have to model the speech variances among different speakers such as dialects or accents of the spoken language, the speaker’s gender to identify what was spoken by the speaker. It is more important in the speech to text conversion systems to convert the accented speech in to text. Telugu language has mainly three different accents namely Coastal Andhra, Rayalaseema and Telangana in which the stress is different for the same word in these accents. In the present research, text dependent speeches from Coastal Andhra, Rayalaseema, Telangana accents have been recorded. The features like pitch, power spectral density, energy and intensity have been extracted and are used to recognize the accent using nearest neighborhood classifier. The best recognition accuracy using this model obtained as 73.3%.
Kasiprasad Mannepalli, P. Narahari Sastry and V. Rajesh, 2015. Accent Detection of Telugu Speech Using Supra-Segmental Features. International Journal of Soft Computing, 10: 287-292.