This dataset was recorded in quiet office and home environments, with the participation of 50 speakers, consisting of 25 males and 25 females. All speakers involved in the recording were carefully selected by professionals 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.