A Unified Deep Neural Network for Speaker and Language Recognition
Computation and Language
2015-04-06 v1 Computer Vision and Pattern Recognition
Machine Learning
Neural and Evolutionary Computing
Machine Learning
Abstract
Learned feature representations and sub-phoneme posteriors from Deep Neural Networks (DNNs) have been used separately to produce significant performance gains for speaker and language recognition tasks. In this work we show how these gains are possible using a single DNN for both speaker and language recognition. The unified DNN approach is shown to yield substantial performance improvements on the the 2013 Domain Adaptation Challenge speaker recognition task (55% reduction in EER for the out-of-domain condition) and on the NIST 2011 Language Recognition Evaluation (48% reduction in EER for the 30s test condition).
Cite
@article{arxiv.1504.00923,
title = {A Unified Deep Neural Network for Speaker and Language Recognition},
author = {Fred Richardson and Douglas Reynolds and Najim Dehak},
journal= {arXiv preprint arXiv:1504.00923},
year = {2015}
}