Automatic Detection Of Voice Disorders Using Self-Supervised Representation Learning Models
Bibliographic record
Abstract
The use of deep neural networks to recognize voice disorders offers good potential for diagnosing these problems, which is of growing interest to researchers and healthcare professionals alike. New findings in self-supervised learning, such as the Wav2vec2, HuBERT, Data2vec and Data2vec2 models, are improving the accuracy and effectiveness of these tools. This research examines their effectiveness in downstream tasks specifically devoted to the automated identification of voice disorders, suggesting a method based on these structures for automatic diagnosis. While Wav2vec2 and HuBERT have already been applied successfully in this domain, in this study, we extend the evaluation by integrating the recent Data2vec models. We focus on a binary classification (healthy versus pathological voices) using the Saarbrücken Voice Database (SVD). Experimental results show that combining the Wav2vec2 model with the CT Transformer classifier delivers outstanding performance: an accuracy rate (ACC) of 86.79 % and an area under the curve (AUC) of 93.3 % for phrase audio samples with 10-fold crossvalidation. These results are among the most highly rated for SVD.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".