rriveranicholas/

dataset_117395161_speech_multimodal3

	dataset_117395161_speech_multimodal3.py




	


	Dataset Summary

A speech dataset with multimodal3 modality, stored in arrow format.

Preprocessing: adaptive Augmentation: autoaugment

Split strategy: kfold 5 Sampling: hard negative

Quality filtering: moderate Labeling: semi auto

dataset_117395161_speech_multimodal3.py — main artifact of… See the full description on the dataset page: https://huggingface.co/datasets/rriveranicholas/dataset_117395161_speech_multimodal3.