This dataset was recorded in both quiet and noisy environments, involving a total of 200 speakers, comprising 101 males and 99 females. All participants in the recording were carefully selected through a professional screening process to ensure standard pronunciation and clarity in speech. The texts for the recordings include information on wake words and other related content.
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.