Multitask Learning with Local Attention for Tibetan Speech Recognition
Open Access
- 18 December 2020
- journal article
- research article
- Published by Wiley in Complexity
- Vol. 2020, 1-10
Abstract
In this paper, we propose to incorporate the local attention in WaveNet-CTC to improve the performance of Tibetan speech recognition in multitask learning. With an increase in task number, such as simultaneous Tibetan speech content recognition, dialect identification, and speaker recognition, the accuracy rate of a single WaveNet-CTC decreases on speech recognition. Inspired by the attention mechanism, we introduce the local attention to automatically tune the weights of feature frames in a window and pay different attention on context information for multitask learning. The experimental results show that our method improves the accuracies of speech recognition for all Tibetan dialects in three-task learning, compared with the baseline model. Furthermore, our method significantly improves the accuracy for low-resource dialect by 5.11% against the specific-dialect model.Keywords
Funding Information
- National Natural Science Foundation of China (61976236)
This publication has 9 references indexed in Scilit:
- Computer-aided diabetic retinopathy diagnostic model using optimal thresholding merged with neural networkInternational Journal of Intelligent Computing and Cybernetics, 2020
- A novel face recognition in uncontrolled environment based on block 2D-CS-LBP features and deep residual networkInternational Journal of Intelligent Computing and Cybernetics, 2020
- End-to-End-Based Tibetan Multitask Speech RecognitionIEEE Access, 2019
- Attention-augmented End-to-end Multi-task Learning for Emotion Prediction from Speech2019
- Multilingual Speech Recognition with a Single End-to-End Model2018
- Joint Modeling of Accents and Acoustics for Multi-Accent Speech Recognition2018
- Multi-task joint-learning of deep neural networks for robust speech recognition2015
- Multi-task Learning of Deep Neural Networks for Low-resource Speech RecognitionIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2015
- Supervised Sequence Labelling with Recurrent Neural Networks2012