|
--- |
|
language: el |
|
datasets: |
|
- aesdd |
|
tags: |
|
- audio |
|
- automatic-speech-recognition |
|
- speech |
|
- xlsr-fine-tuning-week |
|
- speech-emotion-recognition |
|
license: apache-2.0 |
|
--- |
|
|
|
# Emotion Recognition in Greek (el) Speech using Wav2Vec 2.0 |
|
|
|
|
|
## How to use |
|
soon ... |
|
|
|
|
|
## Evaluation |
|
The following tables summarize the scores obtained by model overall and per each class. |
|
|
|
|
|
| Emotions | precision | recall | f1-score | accuracy | |
|
|:---------: |:---------: |:------: |:--------: |:--------: | |
|
| anger | 0.92 | 1.00 | 0.96 | - | |
|
| disgust | 0.85 | 0.96 | 0.90 | - | |
|
| fear | 0.88 | 0.88 | 0.88 | - | |
|
| happiness | 0.94 | 0.71 | 0.81 | - | |
|
| sadness | 0.96 | 1.00 | 0.98 | - | |
|
| - | - | - | - | 91% | |
|
|
|
|
|
## Questions? |
|
Post a Github issue from [HERE](https://github.com/m3hrdadfi/wav2vec/issues). |
