This dataset was recorded in a quiet office environment, with a total of 501 speakers participating, including 239 males and 262 females. All participants involved in the recording were professionally selected to ensure standardized pronunciation and clear enunciation. The recorded texts cover information on command words, wake words, and other related data.