cfahlgren1/crema-d

---
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: actor_id
dtype: string
- name: sentence_code
dtype: string
- name: sentence
dtype: string
- name: emotion_code
dtype: string
- name: emotion_intensity
dtype: string
- name: intensity_code
dtype: string
- name: label
dtype:
class_label:
names:
'0': anger
'1': disgust
'2': fear
'3': happy
'4': neutral
'5': sad
- name: source_file
dtype: string
splits:
- name: train
num_bytes: 674896866
num_examples: 7442
download_size: 605850167
dataset_size: 674896866
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
---

# CREMA-D

Viewer-compatible mirror of the audio portion of CREMA-D (Crowd-sourced Emotional Multimodal Actors Dataset).

This repository republishes the audio files from `myleslinder/crema-d` as Hub-native Parquet shards so the Hugging Face viewer, search, and filtering work without a custom loading script.

## What is included

- `7,442` WAV clips
- `91` actors
- `12` fixed sentence prompts
- `6` emotion labels: `anger`, `disgust`, `fear`, `happy`, `neutral`, `sad`
- `4` intensity codes: `LO`, `MD`, `HI`, `XX`

## Columns

- `audio`: WAV audio at `16000 Hz`
- `actor_id`: actor identifier, for example `1001`
- `sentence_code`: short utterance code, for example `DFA`
- `sentence`: decoded utterance text
- `emotion_code`: short emotion code, for example `ANG`
- `emotion_intensity`: decoded intensity label
- `intensity_code`: original intensity code
- `label`: class label for the emotion
- `source_file`: original filename

## Notes

- This is an audio-only mirror of the Hugging Face dataset `myleslinder/crema-d`.
- The original CREMA-D project also includes video assets, but those files are not part of `myleslinder/crema-d`.
- The original script-based dataset disabled the Hub viewer because it relied on arbitrary Python code. This mirror uses Parquet files instead.

## Source

- Original Hugging Face dataset: `myleslinder/crema-d`
- Original project: <https://github.com/CheyneyComputerScience/CREMA-D>

## Citation

```bibtex
@article{cao2014crema,
title={CREMA-D: Crowd-sourced Emotional Multimodal Actors Dataset},
author={Cao, H. and Cooper, D. G. and Keutmann, M. K. and Gur, R. C. and Nenkova, A. and Verma, R.},
journal={IEEE Transactions on Affective Computing},
volume={5},
number={4},
pages={377--390},
year={2014},
doi={10.1109/TAFFC.2014.2336244}
}
```