Voice Quality Dimensions as Interpretable Primitives for Speaking Style for Atypical Speech and Affect
Not on the map. Why?
Abstract (arXiv)
Perceptual voice quality dimensions describe key characteristics of atypical speech and other speech modulations. Here we develop and evaluate voice quality models for seven voice and speech dimensions (intelligibility, imprecise consonants, harsh voice, naturalness, monoloudness, monopitch, and breathiness). Probes were trained on the public Speech Accessibility (SAP) project dataset with 11,184 samples from 434 speakers, using embeddings from frozen pre-trained models as features. We found that our probes had both strong performance and strong generalization across speech elicitation categories in the SAP dataset. We further validated zero-shot performance on additional datasets, encompassing unseen languages and tasks: Italian atypical speech, English atypical speech, and affective speech. The strong zero-shot performance and the interpretability of results across an array of evaluations suggests the utility of using voice quality dimensions in speaking style-related tasks.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 11 of its 34 references that are in the map. First posted on arXiv 27 May 2025.
Builds on
11 of 34 references.
References · 11
- 2021HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Unitscited by 2,342
- 2019AVEC 2019 Workshop and Challenge: State-of-Mind, Detecting Depression with AI, and Cross-Cultural Affect Recognitioncited by 87
- 2022On the Utility of Self-supervised Models for Prosody-related Taskscited by 42
- 2021Universal Paralinguistic Speech Representations Using Self-Supervised Conformerscited by 32
- 2024ESPnet-SPK: full pipeline speaker embedding toolkit with reproducible recipes, self-supervised front-ends, and off-the-shelf modelscited by 22
- 2021EasyCall corpus: a dysarthric speech datasetcited by 15
- 2023Investigating the Emergent Audio Classification Ability of ASR Foundation Modelscited by 10
- 2023Investigating salient representations and label Variance in Dimensional Speech Emotion Analysiscited by 7
- 2023Seq2seq for Automatic Paraphasia Detection in Aphasic Speechcited by 4
- 2023Speech Intelligibility Classifiers from 550k Disordered Speech Samplescited by 3
- 2024Objective Measurements of Voice Qualitycited by 2
Built on by
Citing papers per year: 2026: 1.
1 paper.