This dataset was recorded in quiet office and home environments, with 20 speakers participating, including 10 males and 10 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear articulation. The recorded texts cover education, cuisine, travel, news, and other information.
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.