Our guest for last week's chat session, Ken Ingham from Amazability talked about his company's efforts to develop a set of speech-based office products. These applications let you use voice -- and voice alone -- for writing, reading, email, web browsing, storing contact information, etc. The applications use middleware for automatic formatting for voice output and to improve voice recognition to up to 100% accuracy for command-and-control.
The primary target for this software is the blind and visually impaired. Ken himself -- the president -- is totally blind. But this software is designed to run on small machines, portables, and even wearable computers. The software is written in native mode. In other words, instead of taking existing office applications and adapting them for voice input and output, the applications are designed from scratch for voice.
As Ken put it, "We do not forget the ink/print forms/styles and have mechanisms to permit the user to learn about them or to specify these during composition or post-composition. However, the key is the efficient imparting of information, which in voice is best done by providing the best possible voicing -- like that in a normal conversation. Such voicing should be done without the need to reference a two-dimensional format or styles."
Ken and his Chief Technical Officer, Peter Olson, are also designing these products based on Linux, rather than Windows, which means much less disk space is needed and the cost of a complete system would be far less. (One of the participants -- Bob Zwick -- pointed out that Fryes Electronics sells Linux systems complete and Internet ready for under $300.)
Ken noted that high end handhelds can support some of Amazability's voice-based applications. But, initially, they are targeting laptops and above.
Longer-term, the possibilities are intriguing.
As Ken put it, "We are replacing the monitor and keyboard with a completely equivalent voice 'console'. Thus, anything that would be displayed on the monitor is sent to our voice output, at which point we filter it."
In other words, these applications -- that enable you to do just about everything you need a computer for -- could run on a computer that had no keyboard and no monitor: a computer that uses voice and voice alone for input and output.
As Bob Fleischer noted, "Voice input and control is considered by
many 'wireless and mobile' people (including myself) to be the
'killer app' or at least the key requirement for the killer app."
As it is today, the smaller the computing gadget, the more awkward
the input and output, because of the dependence on keystrokes and
visual display. Take away that need -- make it easy and effective
to do everything with voice -- and a very powerful computer could
be very small indeed. In addition, you'd no longer have to use up
disk space for software to handle keyboard or stylus input and
visual display. "If a PC can be controlled completely without the
use of a keyboard, we are approaching StarTrek," observed Bob
"A handheld without a screen could be very small indeed," added Bob Fleischer.
Ken emphasized that important problems still remain to be solved. Even with the best fully integrated software, it is easy for a user to get lost, even when things work right. This puts a high burden on the accuracy of the voice recognition and on effective recovery tactics -- like reset buttons.
Basically, by approaching the whole range of common computer applications from the perspective of a blind person -- dropping the assumption that you need a screen or a keyboard -- you can totally avoid design barriers and human usability limits that previously seemed insurmountable.
Seven years ago, I suggested in an article, that in
cyberspace, the blind should be considered as a special resource.
"Companies that want to be on the leading edge in that field
should go out of their way to recruit the blind -- not to conform
to laws about hiring the handicapped and not because it is
politically correct, but rather because their minds are not
totally dominated by visual paradigms. They could imagine, and
with computer technology could simulate, what to the sighted is
unimaginable." The work of Ken Ingham and Amazability seems to be
an instance of that kind of breakthrough.
This site is published by Samizdat