Recherche

Afficher tous les résultats pour

Accueil

Personne

Mathew Magimai Doss

À propos
Confidentialité
Mentions légales

Graph Chatbot

Publications associées (36)

Long-Term Spectral Statistics for Voice Presentation Attack Detection

Mathew Magimai Doss, Sébastien Marcel, Hannah Muckenhirn

Automatic speaker verification systems can be spoofed through recorded, synthetic or voice converted speech of target speakers. To make these systems practically viable, the detection of such attacks, referred to as presentation attacks, is of paramount in ...

2017

A Posterior-Based Multistream Formulation for G2P Conversion

Mathew Magimai Doss, Marzieh Razavi

IEEE Institute of Electrical and Electronics Engineers2017

Acoustic data-driven grapheme-to-phoneme conversion in the probabilistic lexical modeling framework

Mathew Magimai Doss, Ramya Rasipuram, Marzieh Razavi

One of the primary steps in building automatic speech recognition (ASR) and text-to-speech systems is the development of a phonemic lexicon that provides a mapping between each word and its pronunciation as a sequence of phonemes. Phoneme lexicons can be d ...

2016

Articulatory feature based continuous speech recognition using probabilistic lexical modeling

Mathew Magimai Doss, Ramya Rasipuram

Phonological studies suggest that the typical subword units such as phones or phonemes used in automatic speech recognition systems can be decomposed into a set of features based on the articulators used to produce the sound. Most of the current approaches ...

Elsevier2016

Presentation Attack Detection Using Long-Term Spectral Statistics for Trustworthy Speaker Verification

Mathew Magimai Doss, Sébastien Marcel, Hannah Muckenhirn

In recent years, there has been a growing interest in developing countermeasures against non zero-effort attacks for speaker verification systems. Until now, the focus has been on logical access attacks, where the spoofed samples are injected into the syst ...

Ieee2016

Acoustic and Lexical Resource Constrained ASR using Language-Independent Acoustic Model and Language-Dependent Probabilistic Lexical Model

Mathew Magimai Doss, Ramya Rasipuram

One of the key challenges involved in building statistical automatic speech recognition (ASR) systems is modeling the relationship between subword units or “lexical units” and acoustic feature observations. To model this relationship two types of resources ...

2015

On Learning Grapheme-to-Phoneme Relationships through the Acoustic Speech Signal

Mathew Magimai Doss, Ramya Rasipuram

Automatic speech recognition (ASR) systems, through use of the phoneme as an intermediary unit representation, split the problem of modeling the relationship between the written form, i.e., the text and the acoustic speech signal into two disjoint processe ...

2014

Applying Multi- and Cross-Lingual Stochastic Phone Space Transformations to Non-Native Speech Recognition

Hervé Bourlard, Mathew Magimai Doss, David Imseng, John David Scott Dines, Philip Neil Garner

In the context of hybrid HMM/MLP Automatic Speech Recognition (ASR), this paper describes an investigation into a new type of stochastic phone space transformation, which maps "source" phone (or phone HMM state) posterior probabilities (as obtained at the ...

Ieee-Inst Electrical Electronics Engineers Inc2013

Phase AutoCorrelation (PAC) features for noise robust speech recognition

Mathew Magimai Doss, Shajith Ikbal, Hynek Hermansky, Hemant Misra

In this paper, we introduce a new class of noise robust features derived from an alternative measure of autocorrelation representing the phase variation of speech signal frame over time. These features, referred to as Phase AutoCorrelation (PAC) features i ...

2012

A Fast Parts-based Approach to Speaker Verification using Boosted Slice Classifiers

Mathew Magimai Doss, Sébastien Marcel, Anindya Roy

Speaker verification on portable devices like smartphones is gradually becoming popular. In this context, two issues need to be considered: 1) such devices have relatively limited computation resources, and 2) they are liable to be used everywhere, possibl ...

2012

Analysis of MLP Based Hierarchical Phoneme Posterior Probability Estimator

Hervé Bourlard, Mathew Magimai Doss, Hynek Hermansky, Joel Praveen Pinto

We analyze a simple hierarchical architecture consisting of two multilayer perceptron (MLP) classifiers in tandem to estimate the phonetic class conditional probabilities. In this hierarchical setup, the first MLP classifier is trained using standard acous ...

2011

Transcribing Mandarin Broadcast Speech Using Multi-Layer Perceptron Acoustic Features

Mathew Magimai Doss, Fabio Valente, Wen Wang

Recently, several multi-layer perceptron (MLP)- based front-ends have been developed and used for Mandarin speech recognition, often showing significant complementary properties to conventional spectral features. Although widely used in multiple Mandarin s ...

2011

Current trends in multilingual speech processing

Hervé Bourlard, Mathew Magimai Doss, David Imseng, Petr Motlicek, Hui Liang, Lakshmi Babu Saheer, John David Scott Dines, Fabio Valente, Philip Neil Garner

In this paper, we describe recent work at Idiap Research Institute in the domain of multilingual speech processing and provide some insights into emerging challenges for the research community. Multilingual speech processing has been a topic of ongoing int ...

2011

Privacy-Sensitive Audio Features for Speech/Nonspeech Detection

Hervé Bourlard, Mathew Magimai Doss, Daniel Gatica-Perez, Sree Hari Krishnan Parthasarathi

The goal of this paper is to investigate features for speech/nonspeech detection (SND) having low linguistic information from the speech signal. Towards this, we present a comprehensive study of privacy-sensitive features for SND in multiparty conversation ...

2011

Using Comparison of Parallel Phoneme Probability streams for OOV Word Detection

Mathew Magimai Doss, Hynek Hermansky, Tamara Tosic

In this paper, we investigate the approach of comparing two different parallel streams of phoneme posterior probability estimates for OOV word detection. The ﬁrst phoneme posterior probability stream is estimated using only the knowledge of short-term acou ...

2008

Using auxiliary sources of knowledge for automatic speech recognition

Mathew Magimai Doss

Standard hidden Markov model (HMM) based automatic speech recognition (ASR) systems usually use cepstral features as acoustic observation and phonemes as subword units. Speech signal exhibits wide range of variability such as, due to environmental variatio ...

EPFL2005

Page 2 sur 2