Model for Generating On-Off Speech Patterns in Two-Way Conversation
Paul T. Brandy · The Journal of the Acoustical Society of America · 1969
This study presents a model which will generate on-off patterns representative of those obtained in measurements of experimental two-way telephone conversations. The model assumes a conversant to occupy one of three speaking or one of three silent states. Transitions among the states are Markovion, except during certain brief times during which the speaker is forced to remain in his state. Thus, six parameters (one for each state) are required to represent a sneaker. The validity of the model is tested by comparing the patterns generated by the model computer simulation of 16 conversations with patterns obtained from 16 real conversations. Cumulative distribution functions are compared for ten events (such as talkspurts, pauses, mutual silences, etc.) defined on the speech patterns. The model yields good fits to all events except that when an interruption occurs, a model speaker tends to interrupt the other's talkspurt later than a real speaker does. Possible applications of the model include studying people's motivations to talk and fall silent on different circuits, and predicting statistical behavior of voice-operated devices on the circuits.