研究

Table des matières

1 以前的研究主题
- 1.1 博士后
- 1.2 博士
- 1.3 硕士

1 以前的研究主题

1.1 博士后

Speaker Vérification

Participation to the NIST SRE 2008 evaluation campaign Based on the Alizé library developped in the Avignon Laboratory (LIA).

Description of a baseline speaker verification system based on GMM for NIST SRE campaigns : Speaker Verification

Speaker recognition and tracking in audiovisual documents

Attention, in this example, the on screen information is limited to Jacques Chancel as he is the main researched speaker.

Automatic dialogue in real and virtual worlds

Job interview
Distress expression

Involved in several projects

InfoM@gic : pôle de compétitivité Iles-de-France Cap-Digital project, SP5
K-Space : international project, Excellence Network
Myblog3D : ANR project, COST 2102
CompanionAble : international project

Participation to a plenary session of the e-forensic 2009 conference.

负责人

Gérard Chollet (DR CNRS)

实验室

TELECOM ParisTech - TSI (Signal and Image Processing)

1.2 博士

博士论文题目

用于自动语音识别的本地和帧同步置信措施论文题目

博士导师

Jean-Paul Haton (教授) 和 Odile Mella (讲师)

答辩时间

2007年10月9日

论文

论文摘要

In automatic speech recognition, confidence measures aim at estimating the confidence we can give to a result (phone, word, sentence) provided by the speech recognition engine; for example, the contribution of the confidence measure allows to highlight the misrecognized or out-of-vocabulary words.

In this thesis, we propose several confidence measures which are able to provide this estimation for applications using large vocabulary and on-the-fly recognition, as keyword indexation, broadcast news transcription, and live teaching class transcription for hard of hearing children.

In this framework, we have defined two types of confidence measures. The first, based on likelihood ratio, are frame-synchronous measures which can be computed simultaneously with the recognition process of the sentence. The second ones are based on an estimation of the posterior probability limited to a local neighborhood of the considered word, and need only a short delay before being computed on the sub word graph extracted from the recognition process.

These measures were assessed and compared to a state-of-the-art one, which is also based on posterior probability but which requires the recognition of the whole sentence. Two evaluations were performed on a real broadcast news corpus provided by the ESTER campaign. The first one used the Equal Error Rate criterion in an automatic transcription task. The second evaluation was performed in a keyword spotting task. We achieved performance close to our reference measure with our local measures and a delay of less than one second.

We also integrated one of our frame-synchronous measures in the decoding process of the recognition engine in order to improve the solution provided by the system and then to decrease the word error rate. We achieved to decrease the word error rate of around 1%.

Moreover, one of our confidence measure acheived to improve comprehension of hard of hearing children.