Using Prosodic and Acoustic Features for Chinese Dialects Identification
Linjia Sun · 2020
The task of spoken language identification is quite challenging when it comes to discriminating between closely related dialects of the same language. In this paper, a novel approach is proposed to learn and identify the Chinese dialects. We distinguish the difference between Chinese dialects by using the distribution of acoustic and prosodic features, and design to organize these discriminative features into a new topic model. An iterative process is further proposed to implement the model learning. Compared with other state-of-the-art methods, the experimental results show that the proposed model provides competitive performance in the task of Chinese dialects identification.