---
license: cc-by-4.0
tags:
- aggressive
- autoaugment
- contrastive
- image-audio
- speech
- stratified-90-10
- strict
- tfrecord
- weak-supervision
---

# dataset.py

## Dataset Summary

A **speech** dataset with **image audio** modality, stored in **tfrecord** format.

## Preprocessing & Augmentation

- **Preprocessing**: aggressive
- **Augmentation**: autoaugment

## Splits & Sampling

- **Split strategy**: stratified 90 10
- **Sampling**: contrastive

## Quality & Labeling

- **Quality filtering**: strict
- **Labeling**: weak supervision

## Files

- `dataset.py` — main artifact of this repository

## License

See the license field above.