This dataset was recorded in quiet office/home environments, with the participation of 1,018 speakers, including 447 males and 571 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear enunciation. The recorded texts cover information on customer service dialogues, "spot the difference" games, and everyday conversations.
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.
