You could take a look at the Speaker Recognition API from Microsoft Cognitive Services https://www.microsoft.com/cognitive-services/en-us/speaker-r...