Energy Aware Speech Recognition for Mobile Devices

  • Brian Delaney

As portable electronic devices move to smaller form-factors with more features, one challenge is managing and optimizing battery lifetime. Unfortunately, battery technology has not kept up with the rapid pace of semiconductor and wireless technology improvements over the years. In this chapter, we present a study of speech recognition with respect to energy consumption. Our analysis considers distributed speech recognition on hardware platforms with PDA-like functionality. We investigate quality of service and energy trade-offs in this context. We present software optimizations on a speech recognition front-end that can reduce the energy consumption by over 80% compared to the original implementation. A power on/off scheduling algorithm for the wireless interface is presented. This scheduling of the wireless interface can increase the battery lifetime by an order of magnitude. We study the effects of wireless networking and fading channel characteristics on distributed speech recognition using Bluetooth and IEEE 802.11b networks. When viewed as a whole, the optimized distributed speech recognition system can reduce the total energy consumption by over 95% compared to a software client-side ASR implementation. Error concealment techniques can be used to provide further energy savings in low channel SNR conditions.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Acquaviva, A., Simunic, T., Deolalikar, V., and Roy, S. (2003). Remote power control of wireless network interfaces. In Proceedings of PATMOS in Lecture Notes in Computer Science, Springer-Verlag, Turin, September 2003.Google Scholar
  2. Chiasserini, C., Nuggehalli, P., and Srinivasan, V. (2002). Energy-efficient communication protocols. In Proceedings of DAC.Google Scholar
  3. Crenshaw, J. (2000). Math Toolkit for Real-Time Programming. CMP Books, Lawrence, Kansas.Google Scholar
  4. Delaney, B. (2004). Reduced Energy Consumption and Improved Accuracy for Distributed Speech Recognition in Wireless Environments. Ph.D. Thesis, Georgia Institute of Tech-nology.Google Scholar
  5. Delaney, B., Jayant, N., Simunic, T. (2005). Energy-aware distributed speech recognition for wireless mobile devices. IEEE Design and Test of Computers. vol. 22, pp. 39-49.CrossRefGoogle Scholar
  6. Delaney, B. Jayant, N. Hans, M. Simunic, T. Acquaviva, A. (2002). A low-power, fixed-point, front-end feature extraction for a distributed speech recognition system. In Proceedings of ICASSP, pp. I-793-796.Google Scholar
  7. Ebert, J.-P., Aier, S., Kofahl, G., Becker, A., Burns, B., and Wolisz, A. (2002). Measurement and simulation of the energy consumption of a WLAN interface. Tech. Rep. TKN-02-010, Technical University of Berlin, Telecommunication Networks Group.Google Scholar
  8. Frerking, M. E. (1994). Digital Signal Processing in Communications Systems. Van Nostrand Reinhold.Google Scholar
  9. Green, K. and Wilson, J. C. (2001). Future power sources for mobile communications. Elec-tronics and Communication Engineering Journal.Google Scholar
  10. Jones, C., Sivalingam, K., Agrawal, P., and Chen, J. (1999). A survey of energy efficient network protocols for wireless networks. In Proceedings of DATE, pp. 77-81.Google Scholar
  11. Kravets, R. and Krishnan, P. (2000). Application-driven power management for mobile com-munication. Wireless Networks, vol. 6, no. 4, pp. 263-277.MATHCrossRefGoogle Scholar
  12. Krishna, R., Mahlke, S., and Austin, T. (2003). Architectural optimizations for low-power, real-time speech recognition. In Proceedings of CASES.Google Scholar
  13. Lettieri, P., Schurgers, C., and Srivastava, M. (1999). Adaptive link layer strategies for energy efficient wireless networking. Wireless Networks, vol. 5, pp. 339-355.CrossRefGoogle Scholar
  14. Li, X. and Bilmes, J. (2005). Feature pruning for low-power ASR systems in clean and noisy environments. IEEE Signal Processing Letters, vol. 12, no. 7, pp. 489-492.CrossRefGoogle Scholar
  15. Maguire, G. Q., Smith, M., and Beadle, H. W. P. (1998). Smartbadges: A wearable computer and communication system. 6th International Workshop on Hardware/Software Codesign, Invited Talk.Google Scholar
  16. Mathew, B., Davis, A., and Fang, Z. (2003). A Low-power accelerator for the SPHINX 3 speech recognition system. In Proceedings of CASES, 2003.Google Scholar
  17. Nedevschi, S., Patra, R., and Brewer, E. (2005). Hardware speech recognition for user inter-faces in low cost, low power devices. In Proceedings of DAC.Google Scholar
  18. Pearce, D. (2001). Developing the ETSI Aurora advanced distributed speech recognition front-end and what next?. In Proceedings of ASRU, pp. 131-134.Google Scholar
  19. Proakis, J. G. (1995). Digital Communications. McGraw-Hill, 3rd edition.Google Scholar
  20. Shenoy, P. and Radkov, P. (2003) Proxy-assisted power-friendly streaming to mobile devices. In Proceedings of MMCN.Google Scholar
  21. Shih, E., Cho, S., Ickes, N., Min, R., Sinha, A., Wang, A., and Chandrakasan, A. (2001). Physical layer driven protocol and algorithm design for energy efficient wireless sensor networks. In Proceedings of 7th Annual International Conference on Mobile Computing Network.: SIGMOBILE.Google Scholar
  22. Simunic, T., Benini, L., and Micheli, G. D. (2001a). Energy-efficient design of battery-powered embedded systems. IEEE TVLSI (Special Issue), pp. 18-28.Google Scholar
  23. Simunic, T., Benini, L., Glynn, P., and Micheli, G. D. (2001b). Event-driven power manage-ment. IEEE Transactions on CAD.Google Scholar
  24. Smith, P. and Hasler, P.(2002). Analog speech recognition project. In Proceedings of ICASSP, pp. IV-3988-IV-3991.Google Scholar
  25. Valenti, M., Robert, M., and Reed, J. (2002). On the throughput of Bluetooth data transmis-sions. IEEE Wireless Communications and Networking Conference, vol. 1, pp. 119-123.Google Scholar
  26. Wicker, S. B. (1995). Error Control Systems for Digital Communication and Storage. Simon and Schuster.Google Scholar
  27. Zhu, Q. and Alwan, A. (2001). An efficient and scalable 2D DCT-based feature coding scheme for remote speech recognition. In Proceedings of ICASSP.Google Scholar

Copyright information

© Springer-Verlag London Limited 2008

Authors and Affiliations

  • Brian Delaney
    • 1
  1. 1.Lincoln Laboratory, Information Systems Technology GroupMassachusetts Institute of TechnologyLexingtonUSA

Personalised recommendations