This dataset was recorded in a vehicular noise environment, with 283 speakers participating, including 144 males and 139 females. The recorded texts cover information such as numbers, dates, times, and personal names.
Chinese English & American English Speech Recognition Corpus (desktop)
This dataset includes 6 topic for Office White-collar Meeting Scenario - IT and Internet, Finance, Clear Energy, Healthcare, Media and Consumer Electronics.