arXiv subjects

Hossein Zeinali

Publications and source records attributed to Hossein Zeinali.

22 records · Page 2Linked to original sources

arXiv2018

How to Improve Your Speaker Embeddings Extractor in Generic Toolkits

Recently, speaker embeddings extracted with deep neural networks became the state-of-the-art method for speaker verification. In this paper we aim to facilitate its implementation on a more generic toolkit than Kaldi, which we anticipate to enable further improvements on the method. We examine several tricks in training, such as the effects of normalizing input features and pooled statistics, different methods for preventing overfitting as well as alternative non-linearities that can be used instead of Rectifier Linear Units. In addition, we investigate the difference in performance between TDNN and CNN, and between two types of attention mechanism. Experimental results on Speaker in the Wild, SRE 2016 and SRE 2018 datasets demonstrate the effectiveness of the proposed implementation.

cs.SD↗

arXiv2018

Convolutional Neural Networks and x-vector Embedding for DCASE2018 Acoustic Scene Classification Challenge

In this paper, the Brno University of Technology (BUT) team submissions for Task 1 (Acoustic Scene Classification, ASC) of the DCASE-2018 challenge are described. Also, the analysis of different methods on the leaderboard set is provided. The proposed approach is a fusion of two different Convolutional Neural Network (CNN) topologies. The first one is the common two-dimensional CNNs which is mainly used in image classification. The second one is a one-dimensional CNN for extracting fixed-length audio segment embeddings, so called x-vectors, which has also been used in speech processing, especially for speaker recognition. In addition to the different topologies, two types of features were tested: log mel-spectrogram and CQT features. Finally, the outputs of different systems are fused using a simple output averaging in the best performing system. Our submissions ranked third among 24 teams in the ASC sub-task A (task1a).

eess.AS↗

arXiv2018

Spoken Pass-Phrase Verification in the i-vector Space

The task of spoken pass-phrase verification is to decide whether a test utterance contains the same phrase as given enrollment utterances. Beside other applications, pass-phrase verification can complement an independent speaker verification subsystem in text-dependent speaker verification. It can also be used for liveness detection by verifying that the user is able to correctly respond to a randomly prompted phrase. In this paper, we build on our previous work on i-vector based text-dependent speaker verification, where we have shown that i-vectors extracted using phrase specific Hidden Markov Models (HMMs) or using Deep Neural Network (DNN) based bottle-neck (BN) features help to reject utterances with wrong pass-phrases. We apply the same i-vector extraction techniques to the stand-alone task of speaker-independent spoken pass-phrase classification and verification. The experiments on RSR2015 and RedDots databases show that very simple scoring techniques (e.g. cosine distance scoring) applied to such i-vectors can provide results superior to those previously published on the same data.

cs.SD↗

arXiv2017

SUT System Description for NIST SRE 2016

This paper describes the submission to fixed condition of NIST SRE 2016 by Sharif University of Technology (SUT) team. We provide a full description of the systems that were included in our submission. We start with an overview of the datasets that were used for training and development. It is followed by describing front-ends which contain different VAD and feature types. UBM and i-vector extractor training are the next details in this paper. As one of the important steps in system preparation, preconditioning the i-vectors are explained in more details. Then, we describe the classifier and score normalization methods. And finally, some results on SRE16 evaluation dataset are reported and analyzed.

cs.SD↗