Abstract
Call Waiting is a service provided by most telephone companies that alerts a subscriber to an incoming call while he/she is engaged in a prior call. The Talking Call Waiting service at Ameritech enhances Call Waiting by converting Caller ID information into a spoken utterance using text-to-speech technology. A subscriber to Talking Call Waiting hears the name associated with the line that originates a call to them while he/she is on the phone.
Customer acceptance of hearing a text-to-speech synthesized spoken name that interrupts an ongoing conversation was a concern during the concept and design phases of this application. We designed a set of experiments that enabled us to predict customer acceptance of the product based on three factors: 1) the tolerance of the subscriber for interruption in his/her current conversation; 2) the intelligibility of the text-to-speech synthesis (i.e., how successful subscribers are at understanding the name of the call waiting caller); and 3) the perceived quality of the text-to-speech synthesis. We also designed a process that formats the name data for optimal text-to-speech synthesis and a mechanism to respond to possible customer dissatisfaction with the synthesis of particular names. Our research with the service prototype and field system indicate that intelligibility of names was good, quality of the sound was acceptable, and disruption of the call in progress was tolerable. The value of the information about the calling party, conveyed via the spoken name, we believe outweighed the fact that the sound of the speech lacked the quality some customers expected.
Similar content being viewed by others
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Bossemeyer, R., Hardzinski, M. Talking Call Waiting: An Application of Text-to-Speech. International Journal of Speech Technology 4, 7–17 (2001). https://doi.org/10.1023/A:1009637930495
Issue Date:
DOI: https://doi.org/10.1023/A:1009637930495