No abstract

No abstract

To date, speech recognition technology for majority languages has been applied in wireless communication devices successfully. However, as a minority language, Tibetan has very limited resources for conventional automatic speech recognition. It lacks of enough data, sub-word units, lexicons, and word inventories for some dialects. In this paper, we present a multitask end-to-end model to perform simultaneous Tibetan speech content recognition, dialect identification, and speaker recognition. This model avoids processing the pronunciation dictionary and word segmentation for new dialects while allowing for training three tasks in a single model. We build the multitask recognition framework based on WaveNet-CTC. The dialect information and speaker ID are used in the output for training. The experimental results show that our method has better performance compared with a task-specific model. INDEX TERMS End-to-end model, multitask speech recognition, Tibetan language, wavenet model.

Because of ear's special physiological structure and location, it is reasonable to combine ear with profile face for recognition in such scenarios as frontal face images are not available. In this paper, a novel method of multimodal recognition fusing ear and profile face based on kernel principal component analysis is proposed. With the algorithm, the fusion feature vectors of ear and profile face are established and nonlinear feature fusion projection could be implemented. The experimental results show that multimodal recognition fusing both ear and profile face results in improvement over either ear or profile face unimodal recognition.