Martin Geddes

Martin's take on the Future of Conferencing

The Hypervoice team was in Las Vegas March 7-8 to attend the Telespan workshop on the Future of Conferencing. The Telespan event is where the teleconferencing industry gathers to celebrate its wins and mourn its losses. The players cover conference calling, video conferencing, and newer web-centric entrants.

The general tone of the event was one of “change or perish”. The teleconferencing industry is mature, margins are under pressure, and impending regulatory changes in the USA may transform the cost and revenue model. The number of mentions of WebRTC – both as a threat and opportunity – was notably high. For a long time, the per-minute model of conference calls created an incentive to minimise productivity and keep calls long, rather than to maximise it and make calls short. The consequences of this are now catching up.

I presented an Introduction to Hypervoice along with Hypervoice Consortium co-founder Kelly Fitzsimmons, and Tracy Isacke of Telefónica Digital, who is their Director of Business Development & Investments and based in the Silicon Valley office. The idea of Hypervoice clearly excites and interests a lot of people, and the hard question we keep getting is “where do I begin?”

I would like to highlight three players who attended the Telespan event and who have already begun showing value in the Hypervoice space.



Jay Blazensky is co-founder of VoiceBase, and drew a lot of people to his demo at the Hypervoice table. His company is the “Google of spoken voice”. They were demonstrating the ability to ingest large amounts of podcasts and published spoken audio, automatically do voice to text transcription, and then make the results searchable. The search results highlight the position of each search word (e.g. “Advantages” and “Benefits”) in the spoken audio.

Jay made the point that telcos are missing a trick with recorded voice. It can attract new users, make existing ones sticky (by persisting their data), and recruit new ones via the viral nature of sharing.



The ZipDx service provides a fascinating real-time transcription of audio, augmenting the conversation as you speak, as well as providing a parallel text to be able to return to any portion later. Note the “play” button next to each little piece of transcription that lets you return to that place.

The service is of particular interest to international organisations with multi-lingual needs. The ability to apply different language transcription to each separate audio stream in the mix is new and differentiating.


HarQen, one of the six Hypervoice Consortium founding partners, was also present. They have two products in this space. Their VoiceAdvantage product for HR recruitment allows you to quickly find and screen the best candidates for a job, with voice automation resulting in a 2/3 drop in the cost of recruitment. Their Symposia product is an early example of the kind of rich recording, tagging and workflow integration we expect to spread widely in this space. HarQen President Ane Ohm commented: "There is a need to build partnerships and mindshare for this emerging segment. Hypervoice is a rallying point for voice-enhanced business communications."   You can see Symposia in action during our first Hypervoice virtual event, which we blogged about here

Finally, our best wishes to Telespan workshop organiser Elliot Gold who announced his retirement at the close of the event, after 33 years running his business.