|
--- |
|
license: unknown |
|
dataset_info: |
|
features: |
|
- name: path |
|
dtype: string |
|
- name: audio |
|
dtype: audio |
|
- name: emotion |
|
dtype: string |
|
- name: dataset |
|
dtype: string |
|
splits: |
|
- name: train |
|
num_bytes: 4484334276.16 |
|
num_examples: 11970 |
|
download_size: 1315896899 |
|
dataset_size: 4484334276.16 |
|
configs: |
|
- config_name: default |
|
data_files: |
|
- split: train |
|
path: data/train-* |
|
--- |
|
|
|
This dataset is created from [Speech Emotion Recognition (en)](https://www.kaggle.com/datasets/dmitrybabko/speech-emotion-recognition-en) dataset. |
|
|
|
This dataset includes the 4 most popular datasets in English: Crema, Ravdess, Savee, and Tess, containing a total of over 12,000 .wav audio files. Each of these four datasets includes 6 to 8 different emotional labels. |
|
|
|
It includes the 7 types of emotions contained in speech. |
|
|
|
```python |
|
emotions = ['angry', 'disgust', 'fear', 'happy', 'neutral', 'sad', 'surprise'] |
|
``` |
|
|
|
|
|
|
|
|