README.md

metadata

language:
  - ms
  - en

Transcription Instructions

Convert ASR dataset into speech instructions.

1. Sample from mesolitica/pseudolabel-science-large-v3-timestamp
2. Sample from mesolitica/pseudolabel-malaysian-youtube-whisper-large-v3-timestamp
3. Sample from mesolitica/Malaysian-Emilia
4. Mandarin sample from mesolitica/Extra-Emilia/mandarin_v2
5. Cantonese sample from alvanlii/cantonese-radio
6. Entire dataset from mesolitica/Malaysian-STT-Whisper/extra

Download the dataset

huggingface-cli download \
mesolitica/pseudolabel-science-large-v3-timestamp \
--include "*.zip" \
--repo-type "dataset" \
--local-dir './'

huggingface-cli download \
mesolitica/Transcription-Instructions \
--include "*.zip" \
--repo-type "dataset" \
--local-dir './'

huggingface-cli download \
mesolitica/Malaysian-STT-Whisper \
--include "extra.zip" \
--repo-type "dataset" \
--local-dir './'

huggingface-cli download \
mesolitica/Extra-Emilia \
--include 'mandarin-emilia-v2.z*' \
--repo-type "dataset" \
--local-dir './'

huggingface-cli download \
mesolitica/Cantonese-Radio-Description-Instructions \
--include '*.zip' \
--repo-type "dataset" \
--local-dir './'

Acknowledgement

Special thanks to https://www.sns.com.my and Nvidia for 8x H100 node!