Voice is the means by which human beings most commonly and conveniently communicate. The major source of information for many of us, our computers and the Internet, remains locked behind a keyboard - but not for much longer. Voice Web technology promises to enable access to information on the Internet by voice for many practical tasks within the next two to five years.
Voice Web technology does not appear to have the potential for sudden, dramatic discontinuities in the way we carry out processe. It does, however, have the potential to add an important new means of accessing existing services which will require significant work on the part of service providers. Anyone who does not track this technology closely could be badly caught out.
The potential market for voice Web lies with two sectors: those who need it and those who would like it. The former consists primarily of hands-off situations and would use adequate if not ideal technology; for the latter, cosmetic or lifestyle factors will be paramount.
Part of Hewlett-Packard's research into voice Web consists of constructing scenarios, in which devices incorporating the technology would be used. An indication of the size of the in-vehicle market alone, given appropriate devices and services, can be gleaned from statistics from the AA.
According to the AA, there were 22 million cars registered in the UK in late 1998 (the latest available figures) and 144,000 commercial vehicle, although a significant part of the car population will be using their vehicle for commercial purposes too. The commercial sector is important because any identifiable benefits or savings from the use of voice Web will have an immediate commercial case. Moreover, market take-up should be earlier because adequate rather than preferred technology will serve the purpose.
In-vehicle technology is relevant only when people are actually in their vehicles and probably only when making long journeys. There appear to be no statistics for this but the AA estimates there are 144,000 vehicles every day on the northern sector of the M25, from which reasonable inferences on journey length and vehicles in motion at any one time could be made.
All these figures are for the UK, or parts of it. The worldwide market is clearly huge for any successful deployment of the technology. Moreover, all the major vehicle manufacturers appear to be working on voice customisation - the incorporation of optimal voice detection systems. And this is just one of the scenarios HP is working on.
The two- to five-year timescale is key in this case. Elements of the technology required for voice Web interaction are already available but most need improvement. Also, time is needed for standards to be created and implemented and for Web pages to be produced in voice mark-up language. It's probably unwise to expect anything really noticeable to happen within three years, but after that signficant changes could be discernibly under way.
Numerous question marks have to be set against voice Web technology but almost all relate to timescales rather than viability or potential demand. There can be little doubt that HP is at the leading edge of the technology and will stay there, so any doubts relate to when rather than whether. Realisable benefits from technological development will be incremental rather than discontinuous, so key to exploitation will be positioning and understanding of opportunities as they occur.
If all this works to plan, how does HP gain? HP is in the fortunate position of being a seller of base technology, hardware devices and services. It is arguably the major IT supplier that has done most for the creation, implementation and promulgation of open standards (think of printer interfaces and Unix, for instance). It should therefore have little difficulty in selling a good implementation of the base technology. It also has a considerable presence in various hardware markets, making the company well positioned to exploit the market for voice Web devices.