Will AI based IVRs will soon be doomed?

Him Bhankar
petaByts
2 min readJul 31, 2017

--

With the rapid R&D of cloud based AI services from Google, Microsoft, IBM Watson and many other players who offer ground breaking AI services on the cloud, it has now become quite easy to utilise these AI services in different segments.

One of the segments in which AI has recently been used to fullest is AI based IVR systems. With these intelligent IVR systems, a customer does not need to waste time in listening and following IVR’s lame instructions and press different keys, in order to route his problem or query to a specific department. With the advancement of AI’s natural language speech recognition services, most of the customer services are now already using or shifting towards AI based IVR solutions.

Last week, I called my bank’s customer support, they are using AI based IVR and the system is able to recognize and greet me just by listening to my voice. My call got routed to a specific customer care department in few seconds. It felt like a techno-euphoric moment for me.

Recently, I came to know about this amazing AI startup “Lyrebird” based in Montreal. They have taken the wraps off a voice imitation algorithm that the team says can not only mimic the speech of a real person but shift its emotional cadence — and do all this with just a tiny snippet of real world audio. Now, you can copy the voice of anyone. Record 1 minute from someone’s voice and Lyrebird can compress her/his voice’s DNA into a unique key.

This technology will be available to the public as per their latest media announcement so that everyone is aware that this technology exists and that audio recordings are not as reliable as we may think. It is similar to what Photoshop did.

They also put few samples on their website “Lyrebird”, these samples actually mimic President Trump and few other famous politicians.

If you liked this article, click the💚 below so other people will see it here on Medium.

--

--

Him Bhankar
petaByts

Love to work on distributed computing and designing highly available systems.