Robust Several-Speaker Speech Recognition with Highly Dependable Online Speaker Adaptation and Identification

TFIR > Department of Electronics Engineering and Computer Science > journal > Item 987654321/1918

Please use this identifier to cite or link to this item: http://163.15.40.127/ir/handle/987654321/1918

Title:	Robust Several-Speaker Speech Recognition with Highly Dependable Online Speaker Adaptation and Identification
Authors:	Shih, Po-Yi Lin, Po-Chuan Wang, Jhing-Fa Lin, Yuan-Ning 林博川 (東方設計學院電子與資訊系)
Contributors:	東方設計學院電子與資訊系
Keywords:	Speech recognition Speaker adaptation Speaker identification Dependable adaptation Confidence score
Date:	2010-09
Issue Date:	2015-07-14 14:34:23 (UTC+8)
Abstract:	The currently adaptive mechanisms adapt a single acoustic model for a speaker in speaker-independent speech recognition system. However, as more users use the same speech recognizer, single acoustic model adaptation leads to negative adaptation upon switching between users. Such a situation is problematic (undependable adaptation). This paper, considering the situation of a smart home or an office with staff members, presents the speaker-specific acoustic model adaptation based on a multi-model mechanism, to solve the problem of undependable adaptation. First, the identification of the current speaker is confirmed using the SVM classifier, then the corresponding acoustic parameters are extracted and integrated with the speaker-independent acoustic model to yield the speaker-dependent acoustic model and speech recognition accuracy then be promoted for the current speaker. To provide dependable adaptation data to achieve online positive speaker adaptation, a mechanism that measures confidence score is designed to verify each recognition result and determined whether it can be an adaptation datum. The experimental results indicate that the proposed system can effectively increase the average speech recognition accuracy from 62% to 85%. Thus, the proposed system can achieve robust several-speaker speech recognition with highly dependable online speaker adaptation and identification.
Relation:	Journal of Network and Computer Applications, Vol.34 no.5, pp.1459–1467
Appears in Collections:	[Department of Electronics Engineering and Computer Science] journal

Files in This Item:

There are no files associated with this item.

View Licence

Loading...