A Deep Learning Approach Toward Analyzing the Cross‐Lingual Acoustic‐Phonetic Similarities in Multilingual Speech Emotion Recognition
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
This study uses deep learning to explore the influence of phonetic similarities across languages on multilingual SER systems in diverse linguistic contexts. A deep convolutional neural network (DCNN) model was employed to evaluate the performance of speech emotion detection in a multilingual context. Experimented datasets are the SUST Bangla Emotional Speech Corpus (SUBESCO), Indian Institute of Technology Kharagpur Simulated Emotion Hindi Speech Corpus (IITKGP‐SEHSC), SIT Bhubaneswar‐Odia Speech Emotion Database (SITB‐OSED), Ryerson Audio‐Visual Database of Emotional Speech and Song (RAVDESS), and EmoDB datasets of Bangla, Hindi, Odia, English, and German languages, respectively. Here, Bangla, Hindi, and Odia are of the Indo‐Aryan language family and English and German are of the Germanic. A baseline monolingual experiment was performed first to evaluate the models, and then cross‐lingual and multilingual experiments were carried out. The experimental results reveal that the models can recognize emotions of multiple language speech of the same linguistic family better than language speech from different families. The DCNN model achieved the highest multilingual emotion recognition accuracy of 83% for Indo‐Aryan languages, 79% for Germanic languages, and 73% when both language families were combined. These results suggest that phonetic similarities within the same language family improve recognition accuracy.
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.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| 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 | 0.000 |
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 it