Author: Mitchell McLaren

Using bottleneck features extracted from a deep neural network (DNN) trained to predict senone posteriors has resulted in new, state-of-the-art technology for language and speaker identification.

In this work, we explore the role of robust acoustic features motivated by human speech perception studies, for building ASR systems robust to reverberation effects.

This paper compares different approaches for using deep neural networks (DNNs) trained to predict senone posteriors for the task of spoken language recognition (SLR). 

We introduce a new dataset for the study of the effect of highly non-stationary noises on language recognition (LR) performance. 

We recently proposed the use of coefficients extracted from the 2D discrete cosine transform (DCT) of log Mel filter bank energies to improve speaker recognition over the traditional Mel frequency cepstral coefficients (MFCC) with appended deltas and double deltas (MFCC/deltas). 

In this paper we propose softSAD:  the direct integration of speech posteriors into a speaker recognition system instead of using speech activity detection (SAD). 

In this work, we report the same achievement in DNN-based SID performance on microphone speech.  We consider two approaches to DNN-based SID:  one that uses the DNN to extract features, and another that uses the DNN during feature modeling.

This paper applies a convolutional neural network (CNN) trained for automatic speech recognition (ASR) to the task of speaker identification (SID).

This paper explores in depth a recently proposed approach to spoken language recognition based on the estimated posteriors for a set of senones representing the phonetic space of one or more languages. 

We recently proposed the use of deep neural networks (DNN) in place of Gaussian Mixture models (GMM) in the i-vector extraction process for speaker recognition. 

This work presents Trial-Based Calibration (TBC), a novel, automated calibration technique robust to both unseen and widely varying conditions.