The experiments described in this paper are part of a research program on the comprehension of speech produced by text-to-speech synthesizers. In the first experiment, we have measured the reaction times of subjects in recognizing computer generated clicks during the listening of a text read by a synthesizer and by a human speaker. In the second experiment subjects listened to the same passages presented over the telephone channel. This is a difficult listening condition due to the loss of signal quality and to the noise in the channel. The second experiment offers some evidence in favour of the hypothesis that more difficult listening conditions disrupt the performance of listeners of synthetic speech.
Cite as: Delogu, C., Conte, S., Paoloni, A., Sementina, C. (1992) Comprehension of synthetic speech in good and in adverse conditions. Proc. ETRW on Speech Processing in Adverse Conditions, 53-56
@inproceedings{delogu92_spac,
author={Cristina Delogu and S. Conte and A. Paoloni and C. Sementina},
title={{Comprehension of synthetic speech in good and in adverse conditions}},
year=1992,
booktitle={Proc. ETRW on Speech Processing in Adverse Conditions},
pages={53--56}
}