By Heather McLean, 1 October 2001 14:00
NEWS Human beings fool themselves by reading a personality into a synthetic voice even when told it's computer generated. If the synthetic voice mirrors the listeners' personality, the listener is more likely to 'like' it and be influenced by that voice, even to the point of purchasing a product from it. Psychologists at Stanford University in the US assessed how 72 users were influenced by different types of voice reading book reviews from a mock bookstore using different speeds, inflection and intonation. Participants identified the voices as either introvert or extrovert for example, speaking loud and fast. People that called themselves extroverts were attracted to the extroverted voice, the reviews it read plus the reviewer despite constant reminders that a computer was synthesizing the personality, and vice versa for the introverts. According to Doctors' Clifford Nass and Kwan Min Lee, the findings have further meaning for the text to speech system design other than convenience, stating it was also a "rich social modality that must be tuned to the user and the content being presented".
In order to post a comment you need to be registered and logged in.
Log in or create your silicon.com account below