Real-Time Reverb Reduction for Improved Automatic Speech Recognition in Far-Field

Kupryjanow A, Maziewski P, Kurylo L, Lasota P. Real-Time Reverb Reduction for Improved Automatic Speech Recognition in Far-Field. In: AES Convention 142. Audio Engineering Society; 2017. Paper 9803. Available from: https://aes.org/publications/elibrary-page/?id=18677

@inproceedings{Kupryjanow2017_18677,
  author = {Kupryjanow, Adam and Maziewski, Przemyslaw and Kurylo, Lukasz and Lasota, Piotr},
  title = {{Real-Time Reverb Reduction for Improved Automatic Speech Recognition in Far-Field}},
  booktitle = {AES Convention 142},
  note = {Paper 9803},
  year = {2017},
  month = may,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=18677}
}

TY  - CPAPER
TI  - Real-Time Reverb Reduction for Improved Automatic Speech Recognition in Far-Field
AU  - Kupryjanow, Adam
AU  - Maziewski, Przemyslaw
AU  - Kurylo, Lukasz
AU  - Lasota, Piotr
T2  - AES Convention 142
M1  - Paper 9803
PY  - 2017
DA  - 2017/05/06
UR  - https://aes.org/publications/elibrary-page/?id=18677
PB  - Audio Engineering Society
LA  - en
AB  - In the paper, methods of real-time reverb reduction based on Generalized Weighted Prediction Error (GWPE) were presented. It was shown that usage of the proposed audio processing routines highly improve the accuracy of Automatic Speech Recognition (ASR) system namely word error rates (WERs) are reduced 11.36% when the user stands 5 meters from the microphone array. The obtained results are close to the ones that are achieved by the offline GWPE implementation (12.06%). Thanks to optimizations and parameters tuning, computational complexity of the proposed realization of GWPE was highly reduced and it achieves RTFs lower than 1.0 (computation time is shorter than signal duration) when using one core of CPU.
ER  -