Speaker Identification

Speaker Identification is important to the skill I’m creating. The simulator has text input for speaker. Presumably to be able to simulate the ability to recognize the speaker. However, the listen object doesn’t return a speakerId. Is my only option at this point to create rules so the speaker can identify themselves? I’ve checked through the documentation and I see the concept of speaker enrollment, but it seems to be an on-board capability. Hopefully, I didn’t just miss it. Thanks for any insight.


Hi mccune,

When you set up listen rules and get a response back from the speech recognition service, speakerIds is included as a parameter and will return "ID-TI-NOT-AVAILABLE" if you don’t add a speaker name into the simulator. If you do type a speaker name into the simulator, it will return the speakerIds object array. You can access the speakerId by using speakerIds.speakerIds

I hope this helps!