This dataset was recorded in quiet office/home environments, with the participation of 211 speakers, including 102 males and 109 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear enunciation. The recorded texts are of free conversations.