This dataset was recorded in quiet office and home environments, with the participation of 865 speakers, including 490 males and 375 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear enunciation. The recorded texts cover daily spoken language, education, medical health, sports and entertainment, and social economy.