Speech Emotion Recognition Using Hybrid Deep Learning Models
Bibliographic record
Abstract
Human communication conveys emotions not only through words but also through tone, rhythm, and other vocal variations. Accurately identifying these emotions using machines is challenging, especially when relying only on audio input. Speech Emotion Recognition (SER) aims to bridge this gap by analyzing speech signals to determine emotional states. In this work, we present a deep learning-based real-time SER framework that incorporates two data augmentation techniques: background noise injection and spectrogram shifting. The system is trained and tested exclusively on the Toronto Emotional Speech Set (TESS), which contains recordings of female speakers performing different emotions. To extract emotion-relevant cues, multiple acoustic descriptors are used, including Mel-Frequency Cepstral Coefficients (MFCC), chroma features, Root Mean Square (RMS) energy, Zero-Crossing Rate (ZCR), and mel spectrograms. Three neural models were evaluated: a Multilayer Perceptron (MLP), a Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and a hybrid GRU combined with Random Forest (Bi-GRU+RF). Among these, the Bi-GRU+RF with augmentation achieved the highest accuracy, highlighting its effectiveness for speech-based emotion classification. The findings suggest that the proposed system is promising for realtime applications requiring emotion-aware interactions.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 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) | 0.013 | 0.002 |
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; both teacher heads agree on what is shown here.
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".