mesolitica/pseudolabel-malaysian-youtube-whisper-large-v3

---
language:
- ms
task_categories:
- automatic-speech-recognition
---

# Pseudolabel Malaysian Youtube videos using Whisper Large V3

Original dataset at https://huggingface.co/datasets/malaysia-ai/crawl-youtube, distributed pseudolabelled using 4x A100s

script at https://github.com/mesolitica/malaysian-dataset/tree/master/speech-to-text-semisupervised/pseudolabel-whisper

1. Each audio is 30 seconds.
2. Each audio saved in 16k sample rate.

原始信息汇总

数据集概述

语言

• 马来语 (ms)

任务类别

• 自动语音识别 (automatic-speech-recognition)

数据集描述

• 数据集来源于马来西亚YouTube视频，使用Whisper Large V3模型进行伪标签处理。
• 原始数据集链接：malaysia-ai/crawl-youtube
• 伪标签处理脚本链接：mesolitica/malaysian-dataset

数据格式

• 每个音频时长为30秒。
• 每个音频以16k采样率保存。