---
dataset_info:
  features:
  - name: audio
    dtype: audio
  - name: sentence
    dtype: string
  splits:
  - name: train
    num_bytes: 470257191.64
    num_examples: 4060
  download_size: 459865552
  dataset_size: 470257191.64
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
license: cc-by-nc-nd-4.0
language:
- yue
---

ASR-SCCantDuSC: A Scripted Chinese Cantonese (Canton) Daily-use Speech Corpus

This open-source dataset consists of 4.06 hours of transcribed Guangzhou Cantonese scripted speech focusing on daily use sentences, where 4,060 utterances contributed by ten speakers were contained.
Source: https://magichub.com/datasets/guangzhou-cantonese-scripted-speech-corpus-daily-use-sentence/