Skip to content

Detail of publication


Bartoš, T. and Müller, L. : Confidence scoring in ASR using a statistical approach . SPECOM 2005 proceedings, p. 357-360, Moscow State Linguistic University, Moscow , 2005.


Because the ASR systems are not flawless it is desirable to detect the occurred recognition errors already on the phone-level. To cope with this problem, we present a confidence measure (CM) based on our recent experiments. The elementary unit for which the presented CM is investigated is a phone. The first part of this paper shortly introduces the used speech recognition system. The main part of this paper deals with description of our previous experiments and the creation of the new CM based on utilization of synthetically created training data. The performance results of the new CM are also presented in the last part of this paper. A rejection technique based on the described CM was evaluated on the Czech yellow-pages database. Experimental results show that the proposed rejection technique achieves approximately 6% equal error rate (ERR) for phone rejection and about 7-18% EER for word rejection.

Detail of publication

Title: Confidence scoring in ASR using a statistical approach
Author: Bartoš, T. ; Müller, L.
Language: English
Date of publication: 17 Sep 2005
Year: 2005
Type of publication: Papers in proceedings of reviewed conferences
Title of journal or book: SPECOM 2005 proceedings
Page: 357 - 360
ISBN: 5-7452-0110-X
Publisher: Moscow State Linguistic University
Address: Moscow
Date: 17 Sep 2005 - 19 Sep 2005
/ /


confidence measure, statistical approach, hmm


 author = {Barto\v{s}, T. and M\"{u}ller, L.},
 title = {Confidence scoring in ASR using a statistical approach},
 year = {2005},
 publisher = {Moscow State Linguistic University},
 journal = {SPECOM 2005 proceedings},
 address = {Moscow },
 pages = {357-360},
 ISBN = {5-7452-0110-X},
 url = {},