---
dataset_info:
  features:
  - name: audio
    dtype: audio
  - name: sentence
    dtype: string
  splits:
  - name: train
    num_bytes: 1333630236.308
    num_examples: 6219
  download_size: 1335711723
  dataset_size: 1333630236.308
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
language:
- yue
license: cc-by-nc-nd-4.0
---
ASR-SCCantCabSC: A Scripted Chinese Cantonese (Canton) Cabin Speech Corpus

This open-source dataset consists of 5 hours of transcribed Guangzhou Cantonese scripted speech in the vehicle focusing on digits, commands and queries, where 6,219 utterances contributed by ten speakers were contained.
Source: https://magichub.com/datasets/guangzhou-cantonese-scripted-speech-corpus-in-the-vehicle/