SpeechTEK 2013 in New York City this week was a very successful event. I’ve been attending this event for more than 10 years, and while the trade show area was a bit smaller this year than in past years, the quality of attendees, discussions with customers and conference sessions were outstanding.
One of the things I enjoyed most was moderating a panel session on “Assuring Voice Biometrics in the Contact Center.” It was an honor to be on a panel with passionate experts such as Advait Deshpande from Nuance, Ron Delucia from Avaya, Phil Shinn from Morgan Stanley and Dan Miller from Opus Research.
The lively session focused on pragmatic implementation stories about voice biometric deployments. There were many financial institutions in the audience, all with questions, wanting to learn more about this cutting edge technology.
There is no doubt that voice biometric solutions are in early stages. However all of the panelists have seen very early success with voice biometrics, and this trend seems to be continuing. At the same time, there still exists concern about balancing customer experience, privacy and security. These are all valid things to consider, especially when it comes to implementing new technology like voice biometrics.
The session ended all too quickly, but most who were there (and more!) stayed to continue the conversation in the following session presented by Dan Miller on integrating Voice Biometrics.
Learn more: Partnering with Empirix
My Empirix colleague Jason Miller (@motohero) also participated on a panel discussion, and this one was about WebRTC. Bill Scholz from NewSpeech LLC was the panel chair. In addition to Jason, the panel included Valentine Matula from Avaya, Daniel Burnette from Voxeo, and Max Ball from Genesys.
The conversation opened with a general question about what’s going on with WebRTC standards. Being that this event primarily focuses on speech, the group discussed the opportunities around voice biometrics, reducing capital expenditures, and enhancing the customer experience specifically in call centers. Everyone agreed that the first adopters of WebRTC – aside from hardware equipment vendors – would be contact centers.
The panel also touched on who the players in the marketplace are from a hardware and services perspective, and provided opinions about who is launching what and when (Microsoft, Apple, and Google primarily, as well as Avaya, Ericsson, Sansay, and ACME). The final takeaway was that WebRTC is flexible and simple enough for organizations to get started using, even in the state it’s in today.
A Smart Group of Presenters
It was interesting to note that nearly all of the presenters this year work at companies that are Empirix customers. It’s no coincidence. After all, people like talking about successful case studies, and well-planned out implementations that include testing and monitoring are successful.
One of the most interesting keynotes at SpeechTEK was titled “Open the Pod Bay Doors, Siri” and it was presented by Roberto Pieraccini, CEO, International Computer Science Institute. He gave us a history of the speech technology from the view of a speech scientist, discussing how technology has improved, regressed, and might look in the future. His perspective and humor made for a very engaging presentation. And I have to say that I agree with his conclusion: while we’ve come a long way, there’s still much work to do.
Were you at SpeechTEK? Let me know what you learned there in the comments below.
Want to learn more? Check out this paper on predeployment testing methodologies for voice biometrics.
More on WebRTC
2013 has been a year of ginormous communications changes, from the mind-blowing growth of smartphone usage to the game-changing addition of WebRTC. Organizations have been increasingly focusing on improving customer service, recognizing that while voice is still king, customers need to know that they can reach companies in other ways too, like through social media or video chat or instant messaging. Big data has made it possible to begin anticipating issues and improving and personalizing communications.
PESQ is a testing and monitoring related acronym that stands for Perceptual Evaluation of Speech Quality. Typically pronounced “pesk,” it is used as a means for automating the assessment of speech quality in communications networks (Unified Communications, VoIP, etc.).
Still leading the way in terms of the number of customers is Verizon Wireless. With 118 million subscribers, it is just a little ahead of its nearest rival AT&T, who have around 107 million subscribers. In terms of where both were one year ago, this represents only a 2% increase for AT&T but over a 7% increase for Verizon Wireless.
The WebRTC data channel could be used to provide connectivity between the participants in a collaborative session, providing a real-time experience using a peer-to-peer connection.