Robustness of Speaker Recognition from Noisy Speech Samples and Mismatched Languages

Al-Noori A, Li FF, Duncan PJ. Robustness of Speaker Recognition from Noisy Speech Samples and Mismatched Languages. In: AES Convention 140. Audio Engineering Society; 2016. Paper 9577. Available from: https://aes.org/publications/elibrary-page/?id=18275

@inproceedings{AlNoori2016_18275,
  author = {Al-Noori, Ahmed and Li, Francis F. and Duncan, Philip J.},
  title = {{Robustness of Speaker Recognition from Noisy Speech Samples and Mismatched Languages}},
  booktitle = {AES Convention 140},
  note = {Paper 9577},
  year = {2016},
  month = may,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=18275}
}

TY  - CPAPER
TI  - Robustness of Speaker Recognition from Noisy Speech Samples and Mismatched Languages
AU  - Al-Noori, Ahmed
AU  - Li, Francis F.
AU  - Duncan, Philip J.
T2  - AES Convention 140
M1  - Paper 9577
PY  - 2016
DA  - 2016/05/06
UR  - https://aes.org/publications/elibrary-page/?id=18275
PB  - Audio Engineering Society
LA  - en
AB  - Speaker recognition systems can typically attain high performance in ideal conditions. However, significant degradations in accuracy are found in channel-mismatched scenarios. Non-stationary environmental noises and their variations are listed at the top of speaker recognition challenges. Gammtone frequency cepstral coefficient method (GFCC) has been developed to improve the robustness of speaker recognition. This paper presents systematic comparisons between performance of GFCC and conventional MFCC-based speaker verification systems with a purposely collected noisy speech data set. Furthermore, the current work extends the experiments to include investigations into language independency features in recognition phases. The results show that GFCC has better verification performance in noisy environments than MFCC. However, the GFCC shows a higher sensitivity to language mismatch between enrollment and recognition phase.
ER  -