DNN BASED SPEAKER EMBEDDING USING CONTENT INFORMATION FOR TEXT-DEPENDENT SPEAKER VERIFICATION

被引：0

作者：

Dey, Subhadeep ^{[1
,2
]}

Koshinaka, Takafumi ^{[3
]}

Motlicek, Petr ^{[1
]}

Madikeri, Srikanth ^{[1
]}

机构：

[1] Idiap Res Inst, Martigny, Switzerland

[2] Ecole Polytech Fed Lausanne, Lausanne, Switzerland

[3] NEC Corp Ltd, Tokyo, Japan

来源：

2018 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP) | 2018年

关键词：

speaker verification; speaker embedding; i-vectors; content mismatch;

D O I：

暂无

中图分类号：

O42 [声学];

学科分类号：

070206 ; 082403 ;

摘要：

In this paper, we are interested in exploring Deep Neural Network (DNN) based speaker embedding for Random-digit task using content information. To this end, a technique is applied to automatically select common phonetic units between the enrollment and test data to produce speaker verification scores. Furthermore, a novel approach is proposed to incorporate content information in the DNN directly. It is hypothesized that features extracted using this DNN will be helpful for the task. Experiments on the RSR dataset show that the proposed method outperforms the baseline i-vector system by 43% relative equal error rate.

引用

页码：5344 / 5348

页数：5

共 29 条

[1]

[Anonymous], 2011, INTERSPEECH

[2]

[Anonymous], DEEP NEURAL NETWORK

[3] Learning Deep Architectures for AI [J].

Bengio, Yoshua .

FOUNDATIONS AND TRENDS IN MACHINE LEARNING, 2009, 2 (01) :1-127

[4] Current trends in multilingual speech processing [J].

Bourlard, Herve ;

Dines, John ;

Magimai-Doss, Mathew ;

Garner, Philip N. ;

Imseng, David ;

Motlicek, Petr ;

Liang, Hui ;

Saheer, Lakshmi ;

Valente, Fabio .

SADHANA-ACADEMY PROCEEDINGS IN ENGINEERING SCIENCES, 2011, 36 (05) :885-915

[5]

Bredin H, 2017, INT CONF ACOUST SPEE, P5430, DOI 10.1109/ICASSP.2017.7953194

[6]

Chen LP, 2015, 16TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2015), VOLS 1-5, P229

[7]

Collobert R., 2008, P 25 INT C MACH LEAR, P160

[8]

Dey S., 2016, AC SPEECH SIGN PROC

[9] Content Normalization for Text-dependent Speaker Verification [J].

Dey, Subhadeep ;

Madikeri, Srikanth ;

Motlicek, Petr ;

Ferras, Marc .

18TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2017), VOLS 1-6: SITUATED INTERACTION, 2017, :1482-1486

[10]

Dey S, 2017, INT CONF ACOUST SPEE, P5370, DOI 10.1109/ICASSP.2017.7953182

← 1 2 3 →