Analysis and Optimization of Bottleneck Features for Speaker Recognition

Analysis and Optimization of Bottleneck Features for Speaker Recognition

Stať ve sborníku v databázi WoS či Scopus

Recently, Deep Neural Network (DNN) based bottleneck features proved to be very effective in i-vector based speaker recognition. However, the bottleneck feature extraction is usually fully optimized for speech rather than speaker recognition task. In this paper, we explore whether DNNs suboptimal for speech recognition can provide better bottleneck features for speaker recognition. We experiment with different features optimized for speech or speaker recognition as input to the DNN. We also experiment with under-trained DNN, where the training was interrupted before the full convergence of the speech recognition objective. Moreover, we analyze the effect of normalizing the features at the input and/or at the output of bottleneck features extraction to see how it affects the final speaker recognition system performance. We evaluated the systems in the SRE10, condition 5, female task. Results show that the best configuration of the DNN in terms of phone accuracy does not necessary imply better performance of the final speaker recognition system. Finally, we compare the performance of bottleneck features and the standard MFCC features in i-vector/PLDA speaker recognition system. The best bottleneck features yield up to 37% of relative improvement in terms of EER.

Recently, Deep Neural Network (DNN) based bottleneck features proved to be very effective in i-vector based speaker recognition. However, the bottleneck feature extraction is usually fully optimized for speech rather than speaker recognition task. In this paper, we explore whether DNNs suboptimal for speech recognition can provide better bottleneck features for speaker recognition. We experiment with different features optimized for speech or speaker recognition as input to the DNN. We also experiment with under-trained DNN, where the training was interrupted before the full convergence of the speech recognition objective. Moreover, we analyze the effect of normalizing the features at the input and/or at the output of bottleneck features extraction to see how it affects the final speaker recognition system performance. We evaluated the systems in the SRE10, condition 5, female task. Results show that the best configuration of the DNN in terms of phone accuracy does not necessary imply better performance of the final speaker recognition system. Finally, we compare the performance of bottleneck features and the standard MFCC features in i-vector/PLDA speaker recognition system. The best bottleneck features yield up to 37% of relative improvement in terms of EER.

bottleneck features, speaker recognition

bottleneck features, speaker recognition

LOZANO DÍEZ, A.; SILNOVA, A.; MATĚJKA, P.; GLEMBEK, O.; PLCHOT, O.; PEŠÁN, J.; BURGET, L.; GONZALEZ-RODRIGUEZ, J.

2017

21.06.2016

International Speech Communication Association

Bilbao

Proceedings of Odyssey 2016

2312-2846

Proceedings of Odyssey: The Speaker and Language Recognition Workshop Odyssey 2014, Joensuu, Finland

2016

06

Finská republika

352

357

6

http://www.odyssey2016.org/papers/pdfs_stamped/54.pdf

zeinali_interspeech2016_IS161174