Code excited linear predictive coding is one of the more interesting compression techniques for speech. Such low-bit rate stream can be transmitted over the amateur radio frequency bands.
This paper describes a fast algorithm and implementation of code excited linear predictive (CELP) speech coding. It presents principles of the algorithm, including (i) fast conversion of line spectrum pair parameters to linear predictive coding parameters, and (ii) fast searches of the parameters of adaptive and stochastic codebooks. The algorithm can be readily used for speech compression applications, such as on (i) high quality low-bit rate speech transmission in point-to-point or store-and-forward (network based) mode, and (ii) efficient speech storage in speech recording or multimedia databases. The implementation performs in real-time and near real-time on various platforms, including an IBM-PC AT equipped with a TMS320C30 module, an IBM PC 486, a SUN Sparcstation 2, a SUN Sparcstation 5, and an IBM Power PC (Power 590).
A. Langi, VE4ARM, W. Grieder, VE4WSG, and W. Kinsner, VE4WK, "Fast CELP algorithm and implementation for speech compression," in Proceedings of the 13th ARRL Digital Communications Conference (Bloomington, MN; August 19-21, 1994), pp. 97-109, 1994. {ISBN: 0-87259-483-1}
This 13-page paper is a reprint from the proceedings, and can be downloaded as an Adobe PDF file (20 KB). VE4WK is the corresponding author.