This dataset was recorded in both quiet and noisy environments, with a total of 1,012 speakers participating, including 502 males and 510 females. All speakers involved in the recordings were professionally selected to ensure standardized pronunciation and clear articulation. The recorded texts encompass information on wake words, command words, and other related data.
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.
