期刊
IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING
卷 27, 期 1, 页码 77-88出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TASLP.2018.2871755
关键词
Speech recognition; deep neural network (DNN); genetic algorithm; covariance matrix adaptation evolution strategy (CMA-ES); multi-objective optimization
资金
- JSPS KAKENHI [26280055, 17K20001]
- MERL
- Grants-in-Aid for Scientific Research [17K20001, 26280055] Funding Source: KAKEN
The state-of-the-art large vocabulary speech recognition systems consist of several components including hidden Markov model and deep neural network. To realize the highest recognition performance, numerous meta-parameters specifying the designs and training setups of these components must be optimized. A prominent obstacle in system development is the laborious effort required by human experts in tuning these meta-parameters. To automate the process, we propose to tune the meta-parameters of a whole large vocabulary speech recognition system using the evolution strategy with a multi-objective Pareto optimization. As the result of the evolution, the system is optimized for both low word error rate and compact model size. Since the approach requires repeated training and evaluation of the recognition systems that require large computation, we make use of parallel computation on cloud computers. Experimental results show the effectiveness of the proposed approach by discovering appropriate configuration for large vocabulary speech recognition systems automatically.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据