How many unique human voices can Azure Percept Audio recognize at one time.

Question

How many unique human voices can Azure Percept Audio recognize at one time.

vinit sawant 11

If a lot of people are giving commands to percept Audio, can it identify and recognize different voices at a same time?

QuantumCache 20,681 Reputation points Moderator

2022-07-26T06:48:17.767+00:00

Hello @vinit sawant , Just checking back, could you please let us know if you need further help in this matter.
As I am checking with the Engineers internally on the follow-up question, will keep you posted soon.

Please accept the helpful response as 'Answer', which will be helpful to others as well with similar questions.

1 answer

Your answer

QuantumCache 20,681 Reputation points Moderator

2022-07-26T06:48:17.767+00:00

Hello @vinit sawant , Just checking back, could you please let us know if you need further help in this matter.
As I am checking with the Engineers internally on the follow-up question, will keep you posted soon.

Please accept the helpful response as 'Answer', which will be helpful to others as well with similar questions.

Answer 1

QuantumCache 20,681 Moderator

Hello @vinit sawant , Thanks for reaching out with a great question. I hope the below information really helps with your initial query!

The Azure Percept Audio (sometimes called the Percept Ear) is a "System on a Module" or SoM.

The Azure Percept Audio SoM makes use of a couple of Azure Services to process Audio;

LUIS: Allows interaction with applications and devices using natural language.
Cognitive Speech Identification: is an Azure Service offering Text-to-speech, speech-to-text, speech translation and speaker recognition.

Azure Percept Audio datasheet

Please let us know if you need further help in this matter, happy to help!

vinit sawant 11 Reputation points

2022-07-08T07:52:19.827+00:00

Hello @QuantumCache ,

Thank you for your quick response and useful links. That really helped.
But it brings me back to another question which goes like this, If the Audio edge module (The Percept Ear) is placed in a room full of people then how will it identify a keyword spoken by whom or how can it identify different voices like different ID.?

Thanks Again.
QuantumCache 20,681 Reputation points Moderator

2022-07-15T18:00:08.947+00:00

Hello @vinit sawant , let me check on this.