Back to datasets
Dataset assetOpen Source CommunityMental HealthData Analysis

EATD-Corpus

EATD-Corpus is a dataset of audio and text files from 162 volunteers who received counseling. The training set contains data from 83 volunteers (19 depressed and 64 non‑depressed), and the validation set contains data from 79 volunteers (11 depressed and 68 non‑depressed). Each folder contains a volunteer’s depression data, including raw audio, preprocessed audio, audio transcripts, and depression scores.

Source
github
Created
Dec 6, 2021
Updated
Jul 10, 2023
Signals
1,042 views
Availability
Linked source ready
Overview

Dataset description and usage context

EATD-Corpus Dataset Overview

Dataset Description

EATD-Corpus is a dataset of audio and text files composed of 162 volunteers who received counseling.

Download Instructions

The dataset can be downloaded via the following link: Download Link Password: Ymj26Uv5

Usage Instructions

Dataset Split

  • Training Set: Contains data from 83 volunteers (19 depressed and 64 non‑depressed).
  • Validation Set: Contains data from 79 volunteers (11 depressed and 68 non‑depressed).

File Structure

Each folder contains the depression data of one volunteer, with the following files:

  • {positive/negative/neutral}.wav: Raw audio file.
  • {positive/negative/neutral}_out.wav: Preprocessed audio file; preprocessing includes denoising and silence removal.
  • {positive/negative/neutral}.txt: Audio transcript.
  • label.txt: Original SDS score.
  • new_label.txt: Normalized SDS score (original SDS score multiplied by 1.25).
Need downstream help?

Pair the dataset with AI analysis and content workflows.

Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.

Explore AI studio