This is a simple demo of the Microsoft Cognitive Services Speaker Recognition APIs:

1. Speaker Verification
2. Speaker Identification

By using your browser's built-in audio recording capabilities, and a custom version of the fantastic RecorderJS to convert the raw audio into the format required by Cognitive Services, you can easily have a play around with using your voice as a form of verification and also to identify who is talking out of a group of people.

Super cool stuff.

You can view the demo itself here:

• https://rposbo.github.io/speaker-recognition-api/

Articles that support this are over here:

• https://www.robinosborne.co.uk/tag/speech-recognition/

Also you can view the LDN BotFramework Meetup session I gave using this over on YouTube:

• https://youtu.be/dseEDOsWzts?t=28m48s