This dataset was recorded in both quiet and noisy environments, with 201 speakers participating, including 101 males and 100 females. All speakers involved in the recording were professionally selected to ensure standard pronunciation and clear enunciation. The recorded text covers information such as wake words and meetings.
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.