A media company needs to automatically convert thousands of video files containing spoken interviews into text documents for archival and search purposes. The solution should require minimal machine learning expertise and provide time-stamped transcripts. Which AWS service should they use?