This dataset was recorded in a quiet office environment, with the participation of 400 speakers, including 229 males and 171 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear enunciation. The recorded texts cover news and everyday conversations.