Home

Person

Mathew Magimai Doss

About
Privacy
Disclaimer

Graph Chatbot

Related publications (36)

Page 2 of 2

On Modeling Glottal Source Information for Phonation Assessment in Parkinson's Disease

Mathew Magimai Doss, Julian David Fritsch

Parkinson's disease produces several motor symptoms, including different speech impairments that are known as hypokinetic dysarthria. Symptoms associated to dysarthria affect different dimensions of speech such as phonation, articulation, prosody, and inte ...

ISCA-INT SPEECH COMMUNICATION ASSOC2021

Late Fusion of the Available Lexicon and Raw Waveform-based Acoustic Modeling for Depression and Dementia Recognition

Petr Motlicek, Mathew Magimai Doss, Subrahmanya Pavankumar Dubagunta, Julian David Fritsch

Mental disorders, e.g. depression and dementia, are categorized as priority conditions according to the World Health Organization (WHO). When diagnosing, psychologists employ structured questionnaires/interviews, and different cognitive tests. Although acc ...

ISCA-INT SPEECH COMMUNICATION ASSOC2021

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

Using Comparison of Parallel Phoneme Probability streams for OOV Word Detection

Hynek Hermansky, Mathew Magimai Doss, 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

Predicting Heart Activity from Speech using Data-driven and Knowledge-based features

Mathew Magimai Doss, Zohreh Mostaani, Gasser Hassan Elbanna Abdelbaset Abdelrahman Elbanna

Accurately predicting heart activity and other biological signals is crucial for diagnosis and monitoring. Given that speech is an outcome of multiple physiological systems, a significant body of work studied the acoustic correlates of heart activity. Rece ...

ISCA Archives2024

Privacy-Sensitive Audio Features for Speech/Nonspeech Detection

Hervé Bourlard, Daniel Gatica-Perez, Mathew Magimai Doss, 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

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

Sébastien Marcel, Mathew Magimai Doss, 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

Unsupervised Voice Activity Detection by Modeling Source and System Information using Zero Frequency Filtering

Mathew Magimai Doss, Eklavya Sarkar

Voice activity detection (VAD) is an important pre-processing step for speech technology applications. The task consists of deriving segment boundaries of audio signals which contain voicing information. In recent years, it has been shown that voice source ...

ISCA2022

Transcribing Mandarin Broadcast Speech Using Multi-Layer Perceptron Acoustic Features

Mathew Magimai Doss

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

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

Sébastien Marcel, Mathew Magimai Doss, 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

Long-Term Spectral Statistics for Voice Presentation Attack Detection

Sébastien Marcel, Mathew Magimai Doss, 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

Mathew Magimai Doss

Graph Chatbot

On Modeling Glottal Source Information for Phonation Assessment in Parkinson's Disease

Late Fusion of the Available Lexicon and Raw Waveform-based Acoustic Modeling for Depression and Dementia Recognition

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

Using Comparison of Parallel Phoneme Probability streams for OOV Word Detection

Predicting Heart Activity from Speech using Data-driven and Knowledge-based features

Privacy-Sensitive Audio Features for Speech/Nonspeech Detection

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

Unsupervised Voice Activity Detection by Modeling Source and System Information using Zero Frequency Filtering

Transcribing Mandarin Broadcast Speech Using Multi-Layer Perceptron Acoustic Features

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

Long-Term Spectral Statistics for Voice Presentation Attack Detection

A Comparison of Acoustic and Linguistics Methodologies for Alzheimer's Dementia Recognition

On Joint Optimization of Automatic Speaker Verification and Anti-spoofing in the Embedding Space

A Phonology-based Approach for Isolated Sign Production Assessment in Sign Language

Towards Weakly Supervised Acoustic Subword Unit Discovery and Lexicon Development Using Hidden Markov Models

A Posterior-Based Multistream Formulation for G2P Conversion