Shouted and whispered speech compensation for speaker verification systems.

Nowadays, speaker verification systems begin to perform very well under normal speech conditions due to the plethora of neutrally-phonated speech data available, which are used to train such systems. Nevertheless, the use of vocal effort modes other than normal severely degrades performance because of vocal effort mismatch. In this paper, in which we consider whispered, normal and shouted speech production modes, we first study how vocal effort mismatch negatively affects speaker verification performance. Then, in order to mitigate this issue, we describe a series of techniques for score calibration and speaker embedding compensation relying on logistic regression-based vocal effort mode detection. To test the validity of all of these methodologies, speaker verification experiments using a modern x-vector-based speaker verification system are carried out. Experimental results show that we can achieve, when combining score calibration and embedding compensation relying upon vocal effort mode detection, up to 19% and 52% equal error rate (EER) relative improvements under the shouted-normal and whispered-normal scenarios, respectively, in comparison with a system applying neither calibration nor compensation. Compared to our previous work [1] , we obtain a 7.3% relative improvement in terms of EER when adding score calibration in shouted-normal All vs. All condition. • A theoretical study on the vocal effort modes "whispered", "normal" and "shouted" to better understand their main characteristics. • Experiments demonstrate that shouted-normal and whispered-normal vocal effort mismatch drastically decreases speaker verification performance. • Shouted and whispered speech detectors based on logistic regression models that are trained with x-vectors. • Different speaker embedding compensation techniques based on Gaussian mixture models (i.e., MEMLIN, RATZ and SPLICE) in the x-vector domain. • A score calibration technique based on logistic regression models to obtain speaker verification scores robust against vocal-effort variability. [ABSTRACT FROM AUTHOR]