Applying mobile graphics processing unit for speech and behavioral biometrics recognition
Abstract
Mobile graphics processing units (GPU) with low-energy consumption are becoming integral part of mobile devices. Due to widespread of mobile devices research on using mobile GPU for general purpose computations is becoming actual. Research described in our paper is dedicated to use of mobile GPU for computing and time optimization of machine learning algorithms — face and speaker recognition.Our implemented solution proofs that use of mobile GPU considerably lowers energy consumption with simultaneous performance increase comparing to solutions using only central processing unit of conventional mobile devices
Keywords:
speech recognition; voice recognition; grid computing; GPGPU.
References
1. Texas Instruments Inc. ОМАРЗfamily of multimedia application processors. URL: http://focus.ti.com
2. Qualcomm Inc.URL: http://www.qualcomm.com/snapdragon
3. Nvidia Corporation, “Bring High-End Graphics to Handheld Devices,” Nvidia whitepaper, 2011.
4. J. Leskela, J. Nikula, and M. Salmela,“OpenCL Embedded Profile Prototype in Mobile Device," IEEE Workshop on Signal Processing Systems, 2009. PP. 279–284.
5. N. Singhal, I. K. Park, and S. Cho,“Implementation and Optimization of Image Processing Algorithms on Handheld GPU," IEEE International Conference on Image Processing, 2010. PP. 4481–4484.
6. T. Akenine-Moller and J. Strom,“Graphics Processing Units for Handhelds," Proceedings of the IEEE, vol. 96, Issue 5,2008. PP.779–789.
7. J. Strom and T. Akenine-Moller,“iPACKMAN: High-Quality, Low- Complexity Texture Compression for Mobile Phones,” in Proc. Graph. Hardware, 2005. PP. 63–70.
8. Khronos Group, OpenGL ES 2.0 Specification. URL: http://www.khronos. org/opengles.
9. Khronos Group, Open Computing Language (OpenCL) Specification. URL: http://www.khronos.org/opencl.
10. Nvidia Corporation, “Nvidia Compute Unified Device Architecture (CUDA) Programming Guide,” version 2.0,2008.
11. A. Munshi,D. Ginsburg,andD. Shreiner,“OpenGL ES2.0 Programming Guide," Addison-Wesley, USA, 2008.
12. Y. Su, S. Shan, X. Chen, and W. Gao,“Hierarchical Ensemble of Global and Local Classifiers for Face Recognition,” IEEE Transactions on Image Processing, 2009. PP.1885-1896.
13. P. J. Phillips, H. Moon, S.A. Rizvi, andP. J. Rauss, “The FERET Evaluation Methodology for Race-Recognition Algorithms,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000. PP.1090–1104.
14. Open Handset Alliance, Android Software Develop Kit 2.2. URL: http:// developer.android.com/intex.html.
15. T. Sumanaweera and D. Liu,“Medical image reconstruction with the FFT,” In GPU Gems 2, Addison-Wesley, 2005. PP.765–784.
16. D. Ren andR. Suda,“Power Efficient Large Matrices Multiplication by Load Scheduling on Multi-core and GPU Platform with CUDA,” International Conference on Computational Science and Engineering, Vancouver, August
2009.
17. L.D. Brandon, C. Boyd, andN. Govindaraju,“Fast computation of general Fourier Transforms on GPUs,” IEEE International Conference on Multimedia and Expo, Hannover, April 2008.
18. Габдуллин B.B., Капустин А.И., Королев А.И.«Применение технологии CUDAдля задан голосовой биометрии на примере построения универсальной фоновой модели диктора», труды конференции ПаВТ,
2010.
19. Фирун КБ., Лапшин ПА., Фролов В.Е.«Интернет-платформа по работе с аудио-видео информацией с использованием передовых технологий автоматического распознавания речи», труды Vll-ой Конференции молодых учёных НИУ ИТМО, 2011.