The effect of voice input on information exchange in computer supported asynchronous group communication

Computer support for group communication in the business environment is examined. Study of group activities in videotaped business meetings identified information exchange as the most frequent communication task. Existing computer supported cooperative work (CSCW) systems for group communication are typically based on keyboard input, and this may limit their utility. A voice input prototype system for asynchronous (time separated transactions) group communication (AGC) with simulated conversion to text was developed and tested to determine if advantages over conventional keyboard input computer conferencing existed, and whether this method would have any effect on the information exchange task. An experiment utilizing 12 groups of 5 subjects compared the voice input prototype to an existing keyboard input computer conferencing system for an information exchange task, and over all test subjects, voice input resulted in an increase of the number of words contributed by 151% and an increase in the information facts disclosed by 38%. For test subjects holding a high proportion of available facts, voice input resulted in increases of 204% in words and 39% for facts. Effective input speed was 150 words per minute for voice input. These results imply that for future asynchronous group communication systems supporting information exchange, incorporation of voice input capability, coupled with large vocabulary, connected word, speaker independent speech recognition should be considered.