An experiment compared the speaker recognition performance of human listeners to that of computer algorithms/systems. Listening protocols were developed analogous to procedures used in the algorithm evaluation run by the U.S. National Institute of Standards and Technology (NIST), and the same telephone conversation data were used. For "same number" testing, with three-second samples, listener panels and the best algorithm had the same equal-error rate (EER) of 8%. Listeners were better than typical algorithms. For "different number" testing, EER's increased but humans had a 40% lower equal-error rate. Other observations on human listening performance and robustness to "degradations" were made.
Cite as: Schmidt-Nielsen, A., Crystal, T.H. (1998) Human vs. machine speaker identification with telephone speech. Proc. 5th International Conference on Spoken Language Processing (ICSLP 1998), paper 0148, doi: 10.21437/ICSLP.1998-235
@inproceedings{schmidtnielsen98_icslp,
author={Astrid Schmidt-Nielsen and Thomas H. Crystal},
title={{Human vs. machine speaker identification with telephone speech}},
year=1998,
booktitle={Proc. 5th International Conference on Spoken Language Processing (ICSLP 1998)},
pages={paper 0148},
doi={10.21437/ICSLP.1998-235}
}