Language recognition is the task of recognizing the spoken language in a speech utterance. It has applications in multilingual translation systems, automatic speech recognition, targeted advertising, forensics and biometric authentication.
In this project, we aimed to develop a robust language recognition system.
In this paper, we presented a shared view of five institutions resulting from our collaboration towards Language Recognition Evaluation 2015 submissions. We have witnessed a major paradigm shift in adopting deep neural network (DNN) for both feature extraction and classifier.
K. A. Lee, H. Li, L. Deng, V. Hautamäki, W. Rao, X. Xiao, A. Larcher, H. Sun, T. Nguyen, G. Wang, A. Sizov, J. Chen, I. Kukanov, A. H. Poorjam, T. Trong, C. L. Xu, H. H. Xu, B. Ma, E. S. Chng, S. Meignier
in Proc. INTERSPEECH 2016, San Francisco, US, September 2016.
[Paper] & [Technical Report]
In this work, we introduced a new quality measure which represents the uncertainty in estimation of the i-vectors extracted from speech signals. Then, we incorporate it into the recognition process to improve the language recognition accuracy.
Amir H. Poorjam, Rahim Saeidi, Tomi Kinnunen, Ville Hautamäki
in Proc. Speaker and Language Recognition Workshop (Odyssey), pp. 74–80, Bilbao, Spain, June 2016. [PDF]
Presented by Dr. Rahim Saeidi