This study evaluates the effects of some non-learning blind bandwidth extension (BWE) methods on automatic speaker verification (ASV) systems based on x-vector. Recently, a non-linear bandwidth extension (N-BWE) has been proposed as a blind, non-learning, and light-weight BWE approach. Other non-learning BWEs have also been developed in recent years. For ASV evaluations, most data available to train ASV systems is narrowband (NB) telephone speech. Meanwhile, wideband (WB) data have been used to train the state-of-the-art ASV systems, such as i-vector and x-vector. This can cause sampling rate mismatches when all datasets are used. In this paper, we investigate the influence of sampling rate mismatches in the x-vector-based ASV systems and how non-learning BWE methods perform against them. The results showed that the N-BWE method improved the equal error rate (EER) on ASV systems based on x-vector when the mismatches were present. We researched the relationship between objective measurements and EERs. Consequently, the N-BWE method produced the lowest EER and obtained the lower RMS-LSD value and the higher STOI score.
Cite as: Kaminishi, R., Miyamoto, H., Shiota, S., Kiya, H. (2019) Investigation on Blind Bandwidth Extension with a Non-Linear Function and its Evaluation of x-Vector-Based Speaker Verification. Proc. Interspeech 2019, 4055-4059, doi: 10.21437/Interspeech.2019-1510
@inproceedings{kaminishi19_interspeech,
author={Ryota Kaminishi and Haruna Miyamoto and Sayaka Shiota and Hitoshi Kiya},
title={{Investigation on Blind Bandwidth Extension with a Non-Linear Function and its Evaluation of x-Vector-Based Speaker Verification}},
year=2019,
booktitle={Proc. Interspeech 2019},
pages={4055--4059},
doi={10.21437/Interspeech.2019-1510}
}