Publication Details
DNN Based Embeddings for Language Recognition
Plchot Oldřich, Ing., Ph.D. (DCGM)
Matějka Pavel, Ing., Ph.D. (DCGM)
Gonzalez-Rodriguez Joaquin (FIT)
Embeddings, language recognition, LID, DNN
In this work, we present a language identification (LID) system based on embeddings. In our case, an embedding is a fixed-length vector (similar to i-vector) that represents the whole utterance, but unlike i-vector it is designed to contain mostly information relevant to the target task (LID). In order to obtain these embeddings, we train a deep neural network (DNN) with sequence summarization layer to classify languages. In particular, we trained a DNN based on bidirectional long short-term memory (BLSTM) recurrent neural network (RNN) layers, whose frame-by-frame outputs are summarized into mean and standard deviation statistics. After this pooling layer, we add two fully connected layers whose outputs correspond to embeddings. Finally, we add a softmax output layer and train the whole network with multi-class cross-entropy objective to discriminate between languages. We report our results on NIST LRE 2015 and we compare the performance of embeddings and corresponding i-vectors both modeled by Gaussian Linear Classifier (GLC). Using only embeddings resulted in comparable performance to i-vectors and by performing score-level fusion we achieved 7.3% relative improvement over the baseline.
@inproceedings{BUT155045,
author="Alicia {Lozano Díez} and Oldřich {Plchot} and Pavel {Matějka} and Joaquin {Gonzalez-Rodriguez}",
title="DNN Based Embeddings for Language Recognition",
booktitle="Proceedings of ICASSP 2018",
year="2018",
pages="5184--5188",
publisher="IEEE Signal Processing Society",
address="Calgary",
doi="10.1109/ICASSP.2018.8462403",
isbn="978-1-5386-4658-8",
url="https://www.fit.vut.cz/research/publication/11723/"
}