I think it will take me a year to get through all of the companies I visited and interviewed at Oracle OpenWorld.

Voice technology has always been one of my “hobbies”, especially when it comes to the connection into knowledge management or any other Contact Center functionality.  Voice, after all, can be a service just like any other set of functionality within a Services Oriented Architecture. Voice should also be connected to all of the other services available in the enterprise.

Voice-Insight (http://www.voice-insight.com ) was at the show and was demonstrating and talking about some of its technology and some of their R&D projects. Voice-Insight is not necessarily focusing on SOA or services; most of their appeal comes from integration to other technologies.Voice-Insight built its own voice language (VQLTM) that attempts to bridge voice with GIS, maintenance, quality control, vehicle navigation and e-Government applications. This is real; they have  applications running in most West-European languages as well as Japanese and Chinese (Mandarin). Their approach is to try to make user dialogues are close to natural language as possible and minimize user .”Voice training is not required.” Voice-Insight was at the show because they are a big Oracle partner. They also have partnerships with Motorola Inc, IBM, Thales, Magellan Navigation , IBM , ESRI Inc., Acteos (France), Incas (Italy & Spain), Working Machines (USA), ClearOrbit (USA), Patech (UK), MDS (Greece), CAPINFO (China), NBT (Bulgaria), Geodan (The Netherlands), BASF IT (Germany), GTS (RSA) etc.

Voice-Insight Headset

The interesting aspect of Voice-Insight for me is that they are combining what appears to be a robust software and speech offering with hardware. This will afford industries that employ personnel who are not accustomed to using computers the ability to equip their employees with high-tech devices. This will work for companies where you wouldn’t normally think that computers would even survive in the environment; for example refrigerated meat packing areas and extremely dirty environments where employees can’t possibly consider using their hands to push a button or press a key on a computer, laptop, or handheld device.

One of their R&D projects that was interesting to me is SAFIR (Speech Automatic Friendly Interface Research).  The objective of SAFIR is to allow normal people (especially non-computer people) to access, through easy-to-use devices, complex information. For example, giving your Mom easy access to Google Earth and allowing her to move effortlessly through maps while knowing her position via a GPS signal.  The SAFIR project is supported by European Commission and the pilot is running for 48 months and has a budget of about 14M€.

Voice-Insight Applications

Another interesting R&D project underway is PIMAN. It is a research project  financed by the IRSIB (Institute for the Encouragement of Scientific Research and Innovation in the Brussels Region).
Its principal focus is the development of tools and methods in the field of Maintenance and Inspection. The main aim of the PIMAN project is to develop tools and methods that assist an operator in the execution of his/her tasks in the field of Maintenance and Inspection. The specific focus is put on up-to-date information access, assisted data capture, automated positioning, multi-media support, natural language interaction and natural behavior in simulation.  The project is innovative because it uses several advanced technologies in a totally new context. The project combines Voice-Insight’s voice technology with:

Virtual reality: using automated detection of the position of the issue and insertion of the information onto a virtual map of the situation.

Multi-Media: voice, picture, video, etc. in CAD/CAM applied to the field of electronic sub-units and cabling

Machine Learning:  evaluation of contextual information with the aim to improve system performance.

If your environment demands a hands-off approach and your workers don’t have basic computer skills then I would recommend that you take a look at Voice-Insight.