Originally Published MDDI November 2003Product Development Insight

November 1, 2003

2 Min Read
What about two-way communication?

Originally Published MDDI November 2003

Product Development Insight

Michael E. Wiklund
American Institutes for Research

The applications discussed in this article are examples of one-way communication: devices that talk to users but cannot respond to voice commands. An example of this is the automated telephone prompts that airlines use to get callers to speak their origin and destination cities. The technology involves both digital or synthesized speech and speech recognition. It has great promise and is already emerging within the consumer electronics marketplace in products such as cellular phones. However, achieving a reliable level of recognition of commands—close to 100%—continues to be a challenge. Even a recognition rate as high as 95% can cause significant operability problems and user frustration. Getting the systems to work well can seem like a magic act that involves fine-tuning how the system samples, compresses, and buffers the sound data even before they are fed into the recognition software.

To increase reliability, systems may either limit the number of recognized words to a relatively small vocabulary or require users to train the system. Training is accomplished by repeatedly speaking the words to be recognized so that the computer can create a database of spoken words in preparation for determining a match. Despite these limitations, two-way communication technology seems an inevitable next step beyond medical devices that simply talk. One can imagine a wide range of voice-activated medical devices, particularly for people who need their hands for other purposes than pressing buttons and people who have physical disabilities. In fact, some such devices already exist. 

The da Vinci Surgical System from Intuitive Surgical (Sunnyvale, CA) is used to perform minimally invasive bypass surgery on a beating heart. It uses voice activation to activate robots that harvest the arteries used for the bypass. The Socrates Telecollaborative System, also made by Intuitive Surgical, uses voice input to control cameras, lighting, and operating table functions.

Voice activation is also being used to control the movement of an artificial arm developed by students at Johns Hopkins University (Baltimore). The arm responds to the commands raise, down, open, close, and stop. Such applications seem only the beginning in terms of meeting the needs of people with physical impairments.

 Copyright ©2003 Medical Device & Diagnostic Industry

Sign up for the QMED & MD+DI Daily newsletter.

You May Also Like