---
license: other
task_categories:
- automatic-speech-recognition
- audio-classification
language:
- zh
---

The EATD Corpus is hosted in [this github repository](https://github.com/speechandlanguageprocessing/ICASSP2022-Depression).

Follow the instructions there to download and unzip the data.

This dataset can be used with the following line of code, changing the path of `data_dir` to the one appropriate to your system:

```python
dataset = load_dataset('jimregan/eatd_corpus', data_dir='/tmp/EATD-Corpus/')
```