Speech Meets Needs in Many Different Markets
Voice synthesis technology and product identification products and services are developing rapidly and being continuously applied to new markets. While speech-centered companies have focused on the technology for decades, in many industries niches there are unique needs being addressed with speech for the first time.
Speech technology companies such as Compusult Limited are adapting developments and providing low-cost solutions to meet this new demand. This article details some of the solutions available for industries, governments and in the assistive technology field.
Vertical markets present special needs for speech integrators who can provide turnkey and component solutions.
There are for example, a great many problems in the industrial sector which can be solved by speech related technology. These include:
… Orientation systems
… Ways of identifying hazardous material quickly and safely
… "On the fly" reports in high risk professions to reduce dangerous accidents
… Inventory tracking systems from off-site areas to ensure identification of delivery
… Monitoring the integrity of shipments from point of origin through to final delivery
… Vehicle tracking devices
… Adaptation of alarm systems and telephone consoles for visually impaired persons
… Products and services to assist the visually impaired in professional and personal computer environments for greater independence
Tourists and visitors often have a great deal of trouble navigating through new buildings, especially large, not easily accessible locations. Compusult developed NaviTELL, which employs tiny, passive connection-less radio frequency identification (RFID) tags placed at strategic locations within buildings and defined areas, as well as a small, hand-held tag reader capable of activating the radio tags, carried by the user.
The system allows blind users, tourists, visitors and delivery personnel to determine their location, query for direction, monitor their progress and prevent them from getting lost.
Another speech application is helping industry deal with accidents. Until now, authorities have been unable to determine the potential danger in accidents involving commercial vehicles carrying unknown cargo. This results in a slow response time for hazardous clean up.
Compusult has designed and developed an Interrogation and Response Device (IRD) for PAR Government Systems Corporation of New York, which is designed to aid emergency personnel responding to vehicle accidents involving hazardous materials. The IRD is a device which when interrogated by any police radar gun, transmits both speech synthesis output and text data over VHF.
This device allows first responders to an incident to obtain the required vehicle's bill of lading information, including details such as evacuation zones and measures to counter toxic effects while standing back from the incident, without needing any additional or specialized on-board equipment.
Immediate access to this information may help save time, lives and property damage, while potentially reducing clean-up costs and overall environmental damage resulting from an accident.
The IRD integrates commercial off-the-shelf hardware and software, including a speech synthesizer (SounText), a radar detector, a single-board PC, a VHF voice/data transmitter, as well as custom software and circuitry. The IRD will be used to assist in the tracking of hazardous materials and in responding effectively to incidents involving vehicles carrying such materials.
Military Applications
The Canadian Department of National Defence (DND) has also identified new ways to utilize voice synthesis technology in their working environments.
In one example, the DND was faced with the almost impossible task of tracking millions of pieces of armed forces equipment in various locations around the world.
To make their inventory control program efficient and cost effective, while reducing the margin of error, the DND combined pen-based computers, bar-code scanners, programming services with voice synthesis technology for a unique solution.
Up until the spring of this year the Automated Terminal Informa-tion Service (ATIS) messages in Canada consisted only of continuous voice broadcasts of recorded information for arriving and departing aircraft, which were pre-recorded on audio tape by control tower staff using a telephone-like input device.
At that time messages were recorded in a standard format and included weather, types of instrument approach in use, active landing and departing runways and pertinent information on conditions of runways and the serviceability of navigation aids.
To ensure the most accurate information on a timely basis, Compusult designed and developed the ATIS message generator prototype, which includes provision of all commercial and custom software and hardware. The software will run on IBM-compatible PC running Windows 95 or Window NT, and all functions will be accessible within a Windows environment.
When incoming data meet the threshold requirements for creating a new ATIS message, the system will enable a parse engine to retrieve, resolve and process the data, and storing the generated message for subsequent voice synthesis broadcast as an ATIS message.
Operators will be able to examine and edit both raw data and processed messages, determine whether all or part of the incoming data is relevant, and select standard phrases from a pre-defined table for insertion into ATIS messages.
The software will automatically launch the text-to-speech and language translation after the operator has reviewed and approved the ATIS message for release. The ATIS message will then be continuously broadcast to pilots via voice synthesis technology.
Assistive Technology
The world is increasingly a computer-centered place and this is reflected in the growing use of the Internet. For the average person, access to computers and other technologies is readily available, but for the visually impaired, or other individuals with disabilities, access to technology has been difficult in the past.
Accessibility is often limited due to the cost of special equipment such as computers outfitted with screen readers and speech synthesizers or Braille displays and custom input devices. Now, however, speech technology is leveling the playing field with solutions such as:
WebCite, which functions as an enhancement or add-on to commercial Web browsers, incorporates document parsing and interpretation, speech synthesis output and a user interface specifically designed for use by visually and reading impaired persons.
The software integrates custom-developed code with existing commercial off-the-shelf software and is designed to provide support for value-added features which may be incorporated at a later time through standardized interfaces.
ScanTELL identifies and speaks the names of bar-coded items, thus allowing persons who are visually impaired to easily and accurately identify items in their workplaces, homes and retail outlets without assistance.
DigiCite is a portable hand held device which allows visually impaired persons to obtain spoken readouts of numeric LED displays on office and lab equipment, as well as home appliances. The device integrates digital still camera and digital speech technologies in a compact, battery-operated device to provide greater independence and work opportunities for users.
As the demands of the marketplace change, speech technology companies will have to continue to provide innovative solutions to readily address the needs of the various niches within this rapidly expanding field.
With speech technology becoming more pervasive, every market indeed, every industry will find uses for speech recognition as its regular interface between humans and machines. speech can do much more than replace the computer. It has unique advantages that make it the inevitable interface choice for almost all industries and services.