Keywords:
- Acoustic features
- acoustic modeling
- Adaboost
- Alzheimer's disease
- Anti-spoofing
- articulatory features
- Artificial Neural Networks
- ASR
- atypical speech
- audio deepfake
- Automatic accent assessment
- Automatic accent evaluation
- automatic gender recognition
- Automatic speaker verification (ASV)
- Automatic Speech Recognition
- automatic subword unit derivation
- bag of audio words
- bandwidth
- Benchmarking
- Binary features
- binary masking
- bioacoustics
- Blizzard Challenge
- BoAW
- boosting
- breathing pattern estimation
- breathing patterns
- call type classification
- call-type and caller classification
- Children speech recognition
- Classification
- CNN visualization
- ComParE features
- computational efficiency
- Conditional Random Fields
- confidence measures
- continuous speech recognition boosted binary features resource management
- Convolution Neural Network
- Convolutional neural network
- Convolutional Neural Networks
- COVID-19 identification
- cross-database
- cross-transfer knowledge
- Customer satisfaction
- deep learning
- deep neural networks
- deepfake detection
- depression detection
- detection
- Direction of arrival estimation
- discretization
- domain adaptation
- dynamic programming
- Dysarthria
- Dysarthric speech
- embedding
- Emotion Recognition
- emotional prosody
- end-to-end acoustic modeling
- End-to-end learning
- end-to-end modelling
- end-to-end training
- expected performance and spoofability curve
- Expressive Vocalizations
- feature representations
- feature selection
- Few-shot learning
- fine-tuning
- Finetuning
- fixed-size word patterns
- Formant identification
- Formants
- Foundation Model
- Foundation Models
- Fundamental frequency
- Fusion
- Gaussian mixture
- glottal source signals.
- grapheme
- Grapheme subword units
- grapheme subwords
- grapheme-to- phoneme conversion
- grapheme-to-phoneme conversion
- grapheme-to-phoneme converter
- Graphemes
- Hidden Markov Model
- hidden Markov models
- human skeleton estimation
- human speech
- hypoglycemia
- integration of ASV and anti-spoofing
- Inter-pretable Models
- Interpretable features
- isolated word recognition
- Kalman filters
- KL-divergence
- KL-HMM
- Kullback-Leibler divergence
- Kullback-Leibler divergence based hidden Markov model
- Kullback-Leibler divergence based HMM
- Kullback–Leibler divergence based hidden Markov model
- language disorder
- Language Production
- Large Language Models
- leaderboard
- letter-to-sound rules
- lexical model
- Lexical modeling
- Lexicon
- local posterior probability
- localization
- long-term statistics
- LoRA
- low level descriptors
- Low resource language
- machine learning
- Mental Lexicon
- microphone array
- microphone arrays
- mobile biometrics
- modalities fusion
- modified ZFF
- multi- layer perceptron
- Multi-modal Approach
- multi-stream combination
- Multi-task learning
- multilayer perceptron
- multilayer perceptron network
- Multilingual
- multilingual acoustic modeling
- Multimodal
- multiple linear regression
- Multiple speaker localization
- multiple speakers
- multiple-stream combination
- multitask learning
- neural network
- neurocomputational models
- Noise Robustness
- non-native speech
- non-native speech recognition
- noninvasive
- Objective Evaluation
- Objective intelligibility
- Objective intelligibility Assessment
- objective measures
- overlapping speech recognition
- Paralinguistic speech processing
- Parkinson's disease
- Parkinson's disease detection
- Parkinson’s disease
- parts-based approach
- Pathological speech
- Pathological Speech Processing
- PC-GITA
- Peft
- Perceived fluency
- phoneme
- phoneme modeling
- Phoneme recognition
- phoneme subword units
- phoneme subwords
- phonemes
- Phonetic information
- phonetic representation
- Phonocardiogram
- Posterior features
- posterior probabilities
- pre-trained embedding
- pre-training domain
- predictive coding
- presentation attack
- Presentation Attack Detection
- probabilistic lexical modeling
- pronunciation generation
- pronunciation lexicon
- quantization
- Raw Speech
- raw waveform modelling
- raw waveforms
- raw-waveform cnn
- Reading Assessment
- recognition
- recurrent neural network
- Respiratory parameters
- S1-S2 detection
- Scottish Gaelic
- segment-level training.
- Self-Organizing Maps
- Self-supervised embedding
- self-supervised learning
- sentence mode prediction
- sign language assessment
- Sign language processing
- signal processing
- sleepiness
- speaker verification
- speaker-specific features
- spectral statistics
- Speech Analysis
- speech and audio
- speech assessment
- Speech breathing
- Speech Emotion Recognition
- Speech enhancement
- Speech for health
- Speech Foundation Models
- Speech in health
- Speech intelligibility
- speech pathology detection
- speech recognition
- speech recognition.
- speech separation
- speech synthesis
- Speech technology
- Spoken Language Understanding
- Spoofing
- spoofing detection
- Steered response power
- String matching
- SVM
- syllable-level-features
- syllables
- synthetic reference templates.
- Synthetic speech
- TANDEM features
- template-based approach
- template-based system
- Text classification
- text-to-speech synthesis
- token sequences
- tracking
- under-resource speech recognition
- under-resourced languages
- universal phoneme set
- unsupervised adaptation
- utterance verification
- voice
- voice activity detection
- Voice Conversion
- wav2vec2.0
- zero frequency filter
- Zero frequency filtering
- zero-frequency filtering
- zero-resourced speech recognition
- Zero-shot Speech Synthesis
Publications of Mathew Magimai-Doss
2025
| Adaptation of Speech and Bioacoustics Models, , and , Idiap-RR-05-2025 |
|
| Automatic Parkinson’s disease detection from speech: Layer selection vs adaptation of foundation models, , , and , in: Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Hyderabad, India, IEEE, 2025 |
|
| Children's Voice Privacy: First Steps and Emerging Challenges, , , , , and , in: Proceedings of Interspeech, 2025 |
|
| Comparing Self-Supervised Learning Models Pre-Trained on Human Speech and Animal Vocalizations for Bioacoustics Processing, and , in: International Conference on Acoustics, Speech and Signal Processing, 2025 |
|
| Emotion information recovery potential of wav2vec2 network fine-tuned for speech recognition task, and , in: Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Hyderabad, India, IEEE, 2025 |
|
| Exploring the Complexity of Parkinson’s Patient Speech for Depression Detection task: A Qualitative Analysis, , , , and , in: Proceedings of Workshop on Speech Pathology Analysis and DEtection (SPADE), Hyderabad, India, IEEE, 2025 |
|
| Idiap kNN-TTS System for the Blizzard Challenge 2025, , , and , in: Blizzard Challenge Workshop, 2025 |
|
| kNN Retrieval for Simple and Effective Zero-Shot Multi-speaker Text-to-Speech, , , and , in: Proceedings of the Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL), Albuquerque, New Mexico, ACL, 2025 |
[URL] |
| Leveraging Sequential Structure in Animal Vocalizations, and , Idiap-RR-06-2025 |
|
| Listening to Hypoglycemia: Voice as a Biomarker for Detection of a Medical Emergency Using Machine Learning, , , , , , , , , , , and , in: Diabetes Care, 2025 |
[DOI] |
| Measuring negative emotions and stress through acoustic correlates in speech: A systematic review, , , and , in: PLoS One, 20(7), 2025 |
[DOI] |
| Multidisciplinary characterization of embarrassment through behavioral and acoustic modeling, , , , , and , in: Scientific reports, 2025 |
|
| Multimodal Prosody Modeling: A Use Case for Multilingual Sentence Mode Prediction, and , in: Proceedings of Interspeech, 2025 |
|
| On Detection of Depression in Parkinson's Disease Patients' Speech: Handcrafted Features vs. Speech Foundation Models, , , and , in: Automatic Assessment of Parkinsonian Speech, Springer Nature Switzerland AG, 2025 |
[URL] |
| On feature representations for marmoset vocal communication analysis, , , , and , in: Bioacoustics: The International Journal of Animal Sound and its Recording:1-15, 2025 |
[DOI] [URL] |
| Posterior-based analysis of spatio-temporal features for Sign Language Assessment, , , , and , in: IEEE Open Journal of Signal Processing, 2025 |
[DOI] |
| Speech power spectra: a window into neural oscillations in Parkinson's disease, and , Idiap-RR-02-2025 |
|
| Speech power spectra: a window into neural oscillations in Parkinson’s disease, and , in: Proceedings of Interspeech, 2025 |
[DOI] |
| Towards Dynamic Skeleton-based Handshape Subunits for Sign Language Assessment, and , in: Proc. of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2025 |
|
| Towards Leveraging Sequential Structure in Animal Vocalizations, and , in: Neural Information Processing Systems workshop: AI for Non-Human Animal Communication, 2025 |
|
| Unsupervised Rhythm and Voice Conversion of Dysarthric to Healthy Speech for ASR, , , and , in: Proceedings of Workshop on Speech Pathology Analysis and DEtection (SPADE), Hyderabad, India, IEEE, 2025 |
[URL] |
| Unsupervised Rhythm and Voice Conversion to Improve ASR on Dysarthric Speech, , , and , in: Proceedings of Interspeech, Rotterdam, Netherlands, ISCA, 2025 |
[URL] |
| Unveiling Audio Deepfake Origins: A Deep Metric learning And Conformer Network Approach With Ensemble Fusion, , , and , in: Proceedings of Interspeech, 2025 |
|
2024
| COMPARING DATA-DRIVEN AND HANDCRAFTED FEATURES FOR DIMENSIONAL EMOTION RECOGNITION, , and , in: International Conference on Acoustics, Speech and Signal Processing, 2024 |
|
| CONTENT-BASED OBJECTIVE EVALUATION OF ARTIFICIALLY GENERATED SIGN LANGUAGE VIDEOS, , , , , and , in: ICASSP, 2024 |
|
| Cross-transfer Knowledge between Speech and Text Encoders to Evaluate Customer Satisfaction, , , , and , in: Proceedings of Interspeech, Kos Island, Greece, ISCA, 2024 |
|
| Estimating Breathing Pattern from Raw Speech Waveform and Short-term Speech Spectrum using Neural Networks, , , , and , Idiap-RR-12-2024 |
|
| Exploring generalization to unseen audio data for spoofing: insights from SSL models, , , , , and , in: ISCA Proceedings, Greece, 2024 |
[DOI] [URL] |
| Feature Representations for Automatic Meerkat Vocalization Classification, , , and , Idiap-RR-06-2024 |
|
| Feature Representations for Automatic Meerkat Vocalization Classification, , , and , in: 4th International Workshop on Vocal Interactivity in-and-between Humans, Animals and Robots, 2024 |
|
| Neurocomputational model of speech recognition for pathological speech detection: a case study on Parkinson’s disease speech detection, and , in: Proceedings of Interspeech, Kos Island, Greece, pages 3590-3594, 2024 |
[DOI] [URL] |
| On the Utility of Speech and Audio Foundation Models for Marmoset Call Analysis, and , in: 4th International Workshop on Vocal Interactivity in-and-between Humans, Animals and Robots, 2024 |
|
| Posterior-based analysis of spatio-temporal features for Sign Language Assessment, , , , and , Idiap-RR-11-2024 |
|
| Predicting Heart Activity from Speech using Data-driven and Knowledge-based features, , and , in: Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, 2024 |
|
| SSL-TTS: Leveraging Self-Supervised Embeddings and kNN Retrieval for Zero-Shot Multi-speaker TTS, , , and , Idiap-Internal-RR-38-2024 |
[URL] |
| SYLLABLE LEVEL FEATURES FOR PARKINSON'S DISEASE DETECTION FROM SPEECH, and , in: ICASSP, 2024 |
|
| Towards Dynamic Skeleton-based Handshape Subunits for Sign Language Assessment, and , Idiap-RR-09-2024 |
|
| Towards interfacing large language models with ASR systems using confidence measures and prompting, , , , and , in: Proceedings of Interspeech, pages 2980-2984, 2024 |
[DOI] |