This dataset is mainly produced to perform voice cloning task. For voice cloning we need dataset which is basically a audio file, so this file has 
283 voice samples of Barack Obama in .wav format after processing and updating metadata.