This dataset was recorded in quiet office/home environments, with the participation of 415 speakers, including 215 males and 200 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear enunciation. The recorded texts cover news, everyday conversations, and other information.