Fine-tuning the Wav2Vec2 Model for Automatic Speech Emotion Recognition System
Bibliographic record
Abstract
The speech emotion recognition (SER) has gained pivotal attention on various applications in human-computer interaction and affective computing. In these days, there has been a growing interest in developing robust and accurate systems for identifying emotions from speech utterances. In this work, a novel approach to Wav2Vec2 architecture is used to demonstrate the SER system performance. The Wav2Vec2 model is used to extract the speech features from utterances and feed to feed forward network to identify the emotions accurately on the two datasets, namely, Toronto emotional speech set (TESS) and Crowd-sourced Emotional Multimodal Actors Dataset (CREMA-D). Wav2Vec2 implements a contrastive learning target during the pre-training stage. The CREMA-D achieved an accuracy of 76%. Additionally, the weighted F1 score, precision, and recall were, respectively, 0.76, 0.77, and 0.77. On the other hand, on the TESS dataset achieved an accuracy of 99%, the F1 score was 0.99. Furthermore, the weighted recall and precision were both 0.99 and 0.99.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| 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) | 0.003 | 0.003 |
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".