Speech Emotion Recognition in Mental Health: Systematic Review of Voice-Based Applications
Bibliographic record
Abstract
BACKGROUND: The field of speech emotion recognition (SER) encompasses a wide variety of approaches, with artificial intelligence technologies providing improvements in recent years. In the domain of mental health, the links between individuals' emotional states and pathological diagnoses are of particular interest. OBJECTIVE: This study aimed to investigate the performance of tools combining SER and artificial intelligence approaches with a view to their use within clinical contexts and to determine the extent to which SER technologies have already been applied within clinical contexts. METHODS: The review includes studies applied to speech (audio) signals for a select set of pathologies or disorders and only includes those studies that evaluate diagnostic performance using machine learning performance metrics or statistical correlation measures. The PubMed, IEEE Xplore, arXiv, and ScienceDirect databases were queried as recently as February 2025. The Quality Assessment of Diagnostic Accuracy Studies tool was used to measure the risk of bias. RESULTS: A total of 14 articles were included in the final review. The included papers addressed suicide risk (3/14, 21%), depression (8/14, 57%), and psychotic disorders (3/14, 21%). CONCLUSIONS: SER technologies are mostly used indirectly in mental health research and in a wide variety of ways, including different architectures, datasets, and pathologies. This diversity makes a direct assessment of the technology challenging. Nonetheless, promising results are obtained in various studies that attempt to diagnose patients based on either indirect or direct results from SER models. These results highlight the potential for this technology to be used within a clinical setting. Future work should focus on how clinicians can use these technologies collaboratively. TRIAL REGISTRATION: PROSPERO CRD420251006669; https://www.crd.york.ac.uk/PROSPERO/view/CRD420251006669.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.003 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".