You are here:

Artificial Intelligence/Phone Call Voice Recognition Ability.

Advertisement


Question
Dear Chuck

Do the Phone call voice recognition ability varies from person to person ?.

The phone calls received by the person from the calling party has not met the caller face to face.

Example :

John and Dale are phone Caller parties and Peter is receiving phone calls.

1 The Caller party name is John for the first time. He calls up Peter and says "Hi, I am John, I would like to speak to Michael". Peter transfers the phone call to Michael.

2 The Caller party name is Dale for the first time. He calls up Peter and says "Hi, I am Dale, I would like to speak to Michael". Peter transfers the phone call to Michael.

Now after few days, John and Dale again make phone calls, but on phone they do the following conversation. The conversation is as follows.


1 The Caller party name is John. He calls up Peter and says "Hi, I would like to speak to Michael". Will Peter able to recognize the voice of John and without asking John transfers the Line to Michael saying that John is on Phone ?.


2 The Caller party name is Dale. He calls up Peter and says "Hi, I would like to speak to Michael". Will Peter able to recognize the voice of Dale and without asking John transfers the Line to Michael saying that Dale is on Phone ?.


Do the Phone call voice recognition ability will vary from person to person ?. i.e. Peter or somebody who is picking up phone calls. If yes, what are the reasons ?.


Thanks & Regards,
Prashant S Akerkar

Answer
Let me reword your question:  Do modern speech recognition systems learn and adapt to each users specific voice?  The answer to that is an absolute yes.  Dragon speech recognition has been adapting to each user for over 25 years.  Their process has been to have the user 'train' the system to their specific voice patterns by reading standardized text out loud.  This is an especially good way to handle people with different accents or speaking styles.  Dragon has, also, for decades asked permission to read all your documents that you write or work with.  It learns a great deal from this process.  Apple's Siri, Cortana, Google - they all must have some aspect of this, but these systems do not have a specific training process.  However, even Dragon is now claiming to no longer need a training process.  

The specifics of your question well that would be dependent on each specific system.  Some may handle the example you provide, some may not.  

Artificial Intelligence

All Answers


Answers by Expert:


Ask Experts

Volunteer


Chuck Cosby

Expertise

I can answer questions about speech recognition and natural language understanding. I am particulary strong in knowedge based natural langauge techniques. I cannot answer questions about robotics, nueral nets, prolog, or vision recognition - just speech and natural language.

Experience

I have spent 25 years developing natural language software products. I have never developed speech systems, but I have developed sophisticated interfaces from natural language to speech. I have been working with speech recognition systems also for 25 years.

©2016 About.com. All rights reserved.