Salvato in:
Dettagli Bibliografici
Autori principali: Polle, Roseline, Norbury, Agnes, Georgescu, Alexandra Livia, Cummins, Nicholas, Goria, Stefano
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2505.23378
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908389270880256
author Polle, Roseline
Norbury, Agnes
Georgescu, Alexandra Livia
Cummins, Nicholas
Goria, Stefano
author_facet Polle, Roseline
Norbury, Agnes
Georgescu, Alexandra Livia
Cummins, Nicholas
Goria, Stefano
contents Speaker-dependent modelling can substantially improve performance in speech-based health monitoring applications. While mixed-effect models are commonly used for such speaker adaptation, they require computationally expensive retraining for each new observation, making them impractical in a production environment. We reformulate this task as a meta-learning problem and explore three approaches of increasing complexity: ensemble-based distance models, prototypical networks, and transformer-based sequence models. Using pre-trained speech embeddings, we evaluate these methods on a large longitudinal dataset of shift workers (N=1,185, 10,286 recordings), predicting time since sleep from speech as a function of fatigue, a symptom commonly associated with ill-health. Our results demonstrate that all meta-learning approaches tested outperformed both cross-sectional and conventional mixed-effects models, with a transformer-based method achieving the strongest performance.
format Preprint
id arxiv_https___arxiv_org_abs_2505_23378
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Meta-Learning Approaches for Speaker-Dependent Voice Fatigue Models
Polle, Roseline
Norbury, Agnes
Georgescu, Alexandra Livia
Cummins, Nicholas
Goria, Stefano
Machine Learning
Speaker-dependent modelling can substantially improve performance in speech-based health monitoring applications. While mixed-effect models are commonly used for such speaker adaptation, they require computationally expensive retraining for each new observation, making them impractical in a production environment. We reformulate this task as a meta-learning problem and explore three approaches of increasing complexity: ensemble-based distance models, prototypical networks, and transformer-based sequence models. Using pre-trained speech embeddings, we evaluate these methods on a large longitudinal dataset of shift workers (N=1,185, 10,286 recordings), predicting time since sleep from speech as a function of fatigue, a symptom commonly associated with ill-health. Our results demonstrate that all meta-learning approaches tested outperformed both cross-sectional and conventional mixed-effects models, with a transformer-based method achieving the strongest performance.
title Meta-Learning Approaches for Speaker-Dependent Voice Fatigue Models
topic Machine Learning
url https://arxiv.org/abs/2505.23378