This dataset was recorded in a quiet office environment, with 50 speakers participating, including 25 males and 25 females. All speakers who took part in the recording were professionally selected to ensure standardized pronunciation and clear articulation. 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.