Sign In

Communications of the ACM

ACM TechNews

Talking in Color: Voice Imaging Helps Social Skills

View as: Print Mobile App Share:

University of Illinois at Urbana-Champaign computer scientist Karrie Karahalios has developed a technique to digitize conversations and represent them as images, which enables people to "see" their own conversations on computer monitors. Karahalios says the technique provides real-time feedback and can act as a social mirror, allowing people to observe and adjust their own conversations.

The computer program, which Karahalios calls a "conversation clock," presents conversations on a computer terminal as vibrant colors. The colors expand if a voice is talking loudly and overlap when multiple people are talking at the same time. The program has been tested with low-functioning autistic children and in marriage counseling.

people using the 'conversation clock' at the University of Illinois These individuals at the University of Illinois are using the"Conversation Clock." Each colored ring shows whether oneperson is dominating the conversation or whether they aretaking turns. Each voice appears on a computer terminal as a vibrant color. The image grows in size if the voice gets louder, overlaps with another color if it is interrupting, or abruptly narrows with silence. Reuters / Tony Bergstrom / University of Illinois 
It also is being adjusted for use with people with Asperger's Syndrome. People with Asperger's typically possess sophisticated vocabularies but have difficulty with social interactions, often lecturing or monologuing instead of engaging in open conversation. Giving people with Asperger's visual feedback during conversations can help them learn to change their conversational patterns by balancing the colors on the computer screen. Karahalios also has conducted research on using computer interfaces with low-functioning autistic children to get them to be more vocal and use sounds that are the basis of speech, instead of screams or grunts. The system has been successful in getting children to say multiple words.

From Reuters
View Full Article


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account