This dataset was recorded in both quiet office and home environments, with the participation of 500 speakers, including 244 males and 256 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear enunciation. The recorded texts cover wake-up words and other information.