Publication Details
Automatic Language Identification Using Deep Neural Networks
Gonzalez-Dominguez Javier (FIT)
Martínez González David (FIT)
Plchot Oldřich, Ing., Ph.D. (DCGM)
Gonzalez-Rodriguez Joaquin (FIT)
Moreno Pedro (FIT)
Automatic Language Identification, ivectors, DNNs
In this work, we experimented with the use of deep neural networks (DNNs) to automatic language identification (LID). Guided by the success of DNNs for acoustic modelling, we explored their capability to learn discriminative language information from speech signals.
This work studies the use of deep neural networks (DNNs) to address automatic language identification (LID). Motivated by their recent success in acoustic modelling, we adapt DNNs to the problem of identifying the language of a given spoken utterance from short-term acoustic features. The proposed approach is compared to state-of-the-art i-vector based acoustic systems on two different datasets: Google 5M LID corpus and NIST LRE 2009. Results show how LID can largely benefit from using DNNs, especially when a large amount of training data is available. We found relative improvements up to 70%, in Cavg, over the baseline system.
@inproceedings{BUT111548,
author="Ignacio {Lopez-Moreno} and Javier {Gonzalez-Dominguez} and David {Martínez González} and Oldřich {Plchot} and Joaquin {Gonzalez-Rodriguez} and Pedro {Moreno}",
title="Automatic Language Identification Using Deep Neural Networks",
booktitle="Proceeding of ICASSP 2014",
year="2014",
pages="5374--5378",
publisher="IEEE Signal Processing Society",
address="Florencie",
doi="10.1109/ICASSP.2014.6854622",
isbn="978-1-4799-2892-7",
url="http://www.fit.vutbr.cz/research/groups/speech/publi/2014/lopez_moreno_icassp2014_p5374.pdf"
}