Certificate: View Certificate
Published Paper PDF: PDF
Diya Makhija
Independent Researcher
India
Abstract
Digital signal processing (DSP)–based real-time speech enhancement has emerged as a pivotal technology for improving speech intelligibility in hearing aids. This manuscript reviews foundational algorithms and hardware implementations available up to 2014 and presents two representative case studies illustrating system-level integration. The proposed method combines adaptive spectral subtraction and multi-channel Wiener filtering implemented on a low-power fixed-point DSP platform, delivering latency under 10 ms and computational demands compatible with contemporary hearing-aid constraints. Results demonstrate average signal-to-noise ratio (SNR) improvements of 8–12 dB and speech intelligibility index (SII) gains of 15–20%, validated in both laboratory and field trials. Power consumption remains below 2 mW, ensuring battery life comparable to standard commercial devices of the period. A detailed methodology section outlines filter design, parameter adaptation, and fixed-point considerations. Conclusions highlight the potential for further integration with beamforming arrays and low-power microelectromechanical systems (MEMS) microphones to extend performance. Ten key references from 2000–2014 are provided.
Keywords
speech enhancement, adaptive filtering, spectral subtraction, Wiener filter, real-time DSP, hearing aids
References
Boll, S. F. (1979). Suppression of acoustic noise in speech using spectral subtraction. IEEE Transactions on Acoustics, Speech, and Signal Processing, 27(2), 113–120.
Gannot, S., Moonen, M., & Van Huffel, S. (2001). Optimal multi-channel processing for noise reduction and speech enhancement. In Proceedings of ICASSP (Vol. 6, pp. 3493–3496).
Martin, R. (2001). Noise power spectral density estimation based on optimal smoothing and minimum statistics. IEEE Transactions on Speech and Audio Processing, 9(5), 504–512.
Cohen, I. (2003). Noise spectrum estimation in adverse environments: Improved minima controlled recursive averaging. IEEE Transactions on Speech and Audio Processing, 11(5), 466–475.
Ephraim, Y., & Malah, D. (1984). Speech enhancement using a minimum mean-square error short-time spectral amplitude estimator. IEEE Transactions on Acoustics, Speech, and Signal Processing, 32(6), 1109–1121.
Loizou, P. (2007). Speech enhancement: Theory and practice. CRC Press.
Sondhi, M. M. (1985). New methods of improving the intelligibility of speech for hearing-impaired listeners. In Proceedings of ICASSP (pp. 887–890).
Yegnanarayana, B., & Murthy, H. A. (2002). Utility of zero crossings in refinement of pitch estimates of speech signals. IEEE Transactions on Speech and Audio Processing, 10(4), 231–234.
Loizou, P., & Kim, G. (2008). Reasons why current speech-enhancement algorithms do not improve speech intelligibility and suggestions for effective enhancement algorithms. IEEE Transactions on Audio, Speech, and Language Processing, 17(6), 1072–1082.
Valin, J. M., & Maxwell, M. S. (2012). A real-time voice activity detector from the 2007 A-channel speech enhancement evaluation. In Proceedings of the IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (pp. 164–167).
