Attribute Knowledge Integration for Speech Recognition Based on Multi-task Learning Neural Networks

	Attribute Knowledge Integration for Speech Recognition Based on Multi-task Learning Neural Networks
	Hao Zheng1 ; Zhanlei Yang1 ; Liwei Qiao 2; Jianping Li 2; Wenju Liu1
	2015
会议名称	INTERSPEECH
会议录名称	INTERSPEECH
会议日期	2015
会议地点	Dresden, Germany
摘要	It has been demonstrated that the speech recognition performance can be improved by adding extra articulatory information, and subsequently, how to use such information effectively becomes a challenging problem. In this paper, we propose an attribute-based knowledge integration architecture which is realized by modeling and learning both acoustic and articulatory cues simultaneously in a uniform framework. The framework promotes the performance by providing attribute-based knowledge in both feature and model domains. In model domain, the attribute classification is used as the secondary task to improve the performance of an MTL-DNN used for speech recognition by lifting the discriminative ability on pronunciation. In feature domain, an attribute-based feature is extracted from an MTL-DNN trained with attribute classification as its primary task and phonetic/tri-phone state classification as the secondary task. Experiments on TIMIT and WSJ corpuses show that the proposed framework achieves significant performance improvements compared with the baseline DNN-HMM systems.
关键词	Multi-task Learning Automatic Attribute Transcription Deep Neural Networks
收录类别	EI
文献类型	会议论文
条目标识符	http://ir.ia.ac.cn/handle/173211/11779
专题	多模态人工智能系统全国重点实验室_机器人视觉
通讯作者	Hao Zheng
作者单位	1.National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences 2.Electric Power Research Institute of Shanxi Electric Power Company
第一作者单位	模式识别国家重点实验室
通讯作者单位	模式识别国家重点实验室
推荐引用方式 GB/T 7714	Hao Zheng,Zhanlei Yang,Liwei Qiao,et al. Attribute Knowledge Integration for Speech Recognition Based on Multi-task Learning Neural Networks[C],2015.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
IS-2015-1.pdf（388KB）	会议论文		开放获取	CC BY-NC-SA	浏览下载