This dataset was recorded in both quiet office and home environments, with the participation of 460 speakers, including 222 males and 238 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.