SONIVA: Speech recOgNItion Validation in Aphasia
Sanguedolce, G.; Price, C. J.; Brook, S.; Gruia, D. C.; Parkinson, N. V.; Naylor, P. A.; Geranmayeh, F.
Show abstract
Post-stroke aphasia is a major contributor to language impairment and neuro-disability worldwide, making automated assessment a critical research priority. However, the development of clinically validated automatic speech recognition (ASR) systems remains limited by the lack of large, annotated datasets that capture aphasias heterogeneous and unpredictable manifestations. We introduce SONIVA (Speech recOgNItion Validation in Aphasia), the largest and most richly annotated database of pathological speech to date, comprising recordings from {approx}1,000 stroke survivors (including over 200 longitudinally) and {approx}7,000 age-matched controls. Current annotations include 576 patients (mean age: 61.23 {+/-} 13.23 years; 69.81% male) and 104 controls (mean age: 61.05 {+/-} 12.05 years; 34% male), with rich linguistic coding, orthographic and international phonetic alphabet transcriptions. Foundation models finetuned on SONIVA extract linguistic features that correlate with expert transcriptions (Spearmans r = 0.86 - 0.79; p < 0.0001), while acoustic classifiers achieve 93% stroke classification accuracy. These results position SONIVA as a critical resource that can transform rehabilitation through objective, scalable speech assessment.
Matching journals
The top 7 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Hearing and cognitive decline in aging differentially impact neural tracking of context-supported versus random speech across linguistic timescales 93%
- Predictors for Estimating Subcortical EEG Responses to Continuous Speech 92%
- Disentangling listening effort and memory load beyond behavioural evidence 92%
Similar papers in this journal
- Building Large-Scale Registries from Unstructured Clinical Notes using a Low-Resource Natural Language Processing Pipeline 92%
- Deep ensemble multitask classification of emergency medical call incidents combining multimodal data improves emergency medical dispatch 92%
- Uncertainty in Deep Learning for EEG under Dataset Shifts 91%
Similar papers in this journal
Similar papers in this journal
- Digital Voice-Based Biomarker for Monitoring Respiratory Quality of Life: Findings from the Colive Voice Study 94%
- Exploring Language-independent Digital Speech Biomarkers of Hypokinetic Dysarthria 93%
- Unveiling Stimulus Transduction Artifacts in Auditory Steady-State Response Experiments: Characterization, Risks, and Mitigation Strategies 90%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.