Amazon Transcribe allows you to store transcription outputs to your own S3 buckets

Amazon Transcribe offers an automatic speech recognition service that enables to add speech to text capability to the applications by the developers. The audio files stored in the Amazon S3 can be evaluated through Amazon Transcribe API and then have the service return a text file of the transcribed speech. You can generate subtitle for audio and video using Amazon Transcribe which can also be used for transcription of the customer's service calls. Such audio files can be saved or stored in common formats such as MP3 or WAV. Instead of S3 buckets being maintained by the Amazon Transcribe the new update allow you to designate your own S3 buckets to store the transcription outputs. This way the data will be maintained with greater control over the whole transcription workflow. When you create the job you have to first input the name of the designated S3 bucket in the OutputBucketName field and then the Amazon Transcribe will validate the bucket that the bucket exists and it has the permission to write before completing the transcription