This dataset was recorded in a quiet office environment, with the participation of 74 speakers, including 52 males and 22 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear enunciation. The recorded texts cover everyday conversations, news, and command words.