Training "On the Fly" to Improve the Performance of Speaker Recognition in Noisy Environments

Al Noori A, Duncan P, Li F. Training "On the Fly" to Improve the Performance of Speaker Recognition in Noisy Environments. In: AES Conference: 2017 AES International Conference on Audio Forensics. Audio Engineering Society; 2017. Paper 2-2. Available from: https://aes.org/publications/elibrary-page/?id=18744

@inproceedings{AlNoori2017_18744,
  author = {Al Noori, Ahmed and Duncan, Philip and Li, Francis},
  title = {{Training "On the Fly" to Improve the Performance of Speaker Recognition in Noisy Environments}},
  booktitle = {AES Conference: 2017 AES International Conference on Audio Forensics},
  note = {Paper 2-2},
  year = {2017},
  month = jun,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=18744}
}

TY  - CPAPER
TI  - Training "On the Fly" to Improve the Performance of Speaker Recognition in Noisy Environments
AU  - Al Noori, Ahmed
AU  - Duncan, Philip
AU  - Li, Francis
T2  - AES Conference: 2017 AES International Conference on Audio Forensics
M1  - Paper 2-2
PY  - 2017
DA  - 2017/06/06
UR  - https://aes.org/publications/elibrary-page/?id=18744
PB  - Audio Engineering Society
LA  - en
AB  - Reliability of Speaker Recognition (SR) is crucial for critical applications, especially in adverse acoustic conditions. Ambient noises and their variations represent a significant challenge for such applications. In this paper, a new technique is proposed to address the issue of performance degradation in noisy environments. Based on the estimation of the signal to noise ratio (SNR) and profile of the ambient noise from input signals, the proposed method re-trains the enrolment models to generate new noisy models that adapt to the noise profile. This technique is termed “training on the fly”. Evaluation results show notable enhancement in performance in terms of the reduction of equal error rates over a range of SNRs and different types of noise.
ER  -