Google Speech offer speech recognition that can create transcripts that are fairly good as a starting point for transcribing interviews. It will naturally not transcribe the interviews perfectly, however I find it useful enough as a starting point. Instead of manually uploading files using python, would it be possible to create a feature to transcribe interviews in Nvivio using the Google Speach API, MS Cortana or Amazon Alexa?

Google Speech currently offer speech recognition in more languages than MS Cortana and Amazon Alexa, but they will all work i suppose.