Q2: Algorithms and standards

CELP stands for "code excited linear prediction". LPC stands for "linear predictive coding". They are compression algorithms used for low bit rate (2400 and 4800 bps) speech coding.

The U.S. DoD's Federal-Standard-1016 based 4800 bps code excited linear prediction voice coder version 3.2 (CELP 3.2) Fortran and C simulation source codes are available for worldwide distribution (on DOS diskettes, but configured to compile on Sun SPARC stations) from NTIS and DTIC. Example input and processed speech files are included. A Technical Information Bulletin (TIB), "Details to Assist in Implementation of Federal Standard 1016 CELP," and the official standard, "Federal Standard 1016, Telecommunications: Analog to Digital Conversion of Radio Voice by 4,800 bit/second Code Excited Linear Prediction (CELP)," are also available.

To obtain CELP:

Available through the National Technical Information Service:

NTIS
U.S. Department of Commerce
5285 Port Royal Road
Springfield, VA 22161
USA
(800) 553-6847

FS-1016 CELP 3.2 may also be obtained from FS-1016 CELP 3.2 may also be obtained from ftp://svr-ftp.eng.cam.ac.uk/pub/comp.speech/coding/celp_3.2a.tar.Z or ftp://ftp.super.org/pub/speech/celp_3.2a.tar.Z.

LPC is available from ftp://ftp.super.org/pub/speech/lpc10-1.0.tar.gz or ftp://svr-ftp.eng.cam.ac.uk/pub/speech/coding/lpc10-1.0.tar.gz.

MATLAB software for LPC-10 is available from http://www.eas.asu.edu/~spanias/srtcrs.html. Also, postscript copies of tutorials of speech coding can be found at http://www.eas.asu.edu/~spanias/papers.html. [Andreas Spanias, spanias@asu.edu]

For more information:

 

• The following articles describe the Federal-Standard-1016 4.8-kbps CELP coder (it's unnecessary to read more than one):

  Campbell, Joseph P. Jr., Thomas E. Tremain and Vanoy C. Welch, The Federal Standard 1016 4800 bps CELP Voice Coder, Digital Signal Processing, Academic Press, 1991, Vol. 1, No. 3, p. 145-155.

  Campbell, Joseph P. Jr., Thomas E. Tremain and Vanoy C. Welch, The DoD 4.8 kbps Standard (Proposed Federal Standard 1016), in Advances in Speech Coding, ed. Atal, Cuperman and Gersho, Kluwer Academic Publishers, 1991, Chapter 12, p. 121-133.

  Campbell, Joseph P. Jr., Thomas E. Tremain and Vanoy C. Welch, The Proposed Federal Standard 1016 4800 bps Voice Coder: CELP, Speech Technology Magazine, April/May 1990, p. 58-64.

  Additional information on CELP can also be found in the comp.speech FAQ.

• The voicing classifier used in the enhanced LPC-10 (LPC-10e) is described in: Campbell, Joseph P., Jr. and T. E. Tremain, Voiced/Unvoiced Classification of Speech with Applications to the U.S. Government LPC-10E Algorithm, Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing, 1986, p. 473-6.

  The U. S. Federal Standard 1015 (NATO STANAG 4198) is described in: Thomas E. Tremain, The Government Standard Linear Predictive Coding Algorithm: LPC-10, Speech Technology Magazine, April 1982, pp. 40-49.

[Most of the above from Joe Campbell, jpcampb@afterlife.ncsc.mil, with additions from Dan Frankowski, drankow@winternet.com, and Ed Hall, edhall@rand.org]

ADPCM stands for Adaptive Differential Pulse Code Modulation. It is a family of speech compression and decompression algorithms. A common implementation takes 16-bit linear PCM samples samples and converts them to 4-bit samples, yielding a compression rate of 4:1.

To obtain:

There is public domain C code available via anonymous ftp at ftp://ftp.cwi.nl/pub/audio/adpcm.shar written by Jack Jansen (email Jack.Jansen@cwi.nl). It is very programmer-friendly. The ADPCM code used is the Intel/DVI ADPCM code which is being recommended by the IMA Digital Audio Technical Working Group. It allows the following calls: There is public domain C code available via anonymous ftp at ftp://ftp.cwi.nl/pub/audio/adpcm.shar written by Jack Jansen (email Jack.Jansen@cwi.nl). It is very programmer-friendly. The ADPCM code used is the Intel/DVI ADPCM code which is being recommended by the IMA Digital Audio Technical Working Group. It allows the following calls:

adpcm_coder(short inbuf[], char outbuf[], int nsample,
        struct adpcm_state *state);
adpcm_decoder(char inbuf[], short outbuf[], int nsample,
        struct adpcm_state *state);

Note that this is NOT a G.722 coder. The ADPCM standard is much more complicated, probably resulting in better quality sound but also in much more computational overhead.

Platforms:

The routines have been tested on numerous platforms, and will easily compress and decompress millions of samples per second on current hardware.

For more information:

The G.721/722/723 packages are available from ITU at http://www.itu.ch/.

This is also available as: ftp://svr-ftp.eng.cam.ac.uk/pub/comp.speech/coding/G711_G722_G723.tar.gz

[From Dan Frankowski, dfrankow@winternet.com; Jack Jansen, Jack.Jansen@cwi.nl]

GSM is a standard for digital cellular telephony used in Europe. In this context, GSM also refers to the speech coder used in GSM telephones.

The Communications and Operating Systems Research Group (KBS) at the Technische Universitaet Berlin is currently working on a set of UNIX-based tools for computer-mediated telecooperation that will be made freely available.

As part of this effort we are publishing an implementation of the European GSM 06.10 provisional standard for full-rate speech transcoding, prI-ETS 300 036, which uses RPE/LTP (residual pulse excitation/long term prediction) coding at 13 kbit/s.

GSM 06.10 compresses frames of 160 13-bit samples (8 kHz sampling rate, i.e. a frame rate of 50 Hz) into 260 bits; for compatibility with typical UNIX applications, our implementation turns frames of 160 16-bit linear samples into 33-byte frames (1650 Bytes/s). The quality of the algorithm is good enough for reliable speaker recognition; even music often survives transcoding in recognizable form (given the bandwidth limitations of 8 kHz sampling rate).

The interfaces offered are a front end modeled after compress(1), and a library API. Compression and decompression run faster than realtime on most SPARCstations. The implementation has been verified against the ETSI standard test patterns.

Jutta Degener jutta@cs.tu-berlin.de, Carsten Bormann cabo@cs.tu-berlin.de)

Communications and Operating Systems Research Group, TU Berlin
Fax: +49.30.31425156, Phone: +49.30.31424315

To obtain:

ftp://svr-ftp.eng.cam.ac.uk/pub/comp.speech/coding/gsm-1.0.6.tar.gz. An alternative site is ftp://ftp.cwi.nl/pub/audio/gsm-1.0.7.tar.gz. Try also: http://kbs.cs.tu-berlin.de/~jutta/toast.html.

[From Dan Frankowski, dfrankow@winternet.com; Jutta Degener, jutta@cs.tu-berlin.de]

Pitch is officially defined as "That attribute of auditory sensation in terms of which sounds may be ordered on a musical scale." Several good examples illustrating the subtleties of pitch perception are included in the "Auditory Demonstrations CD" which is available from the Acoustical Society of America, Woodbury, NY 10797 for $20.

Books/papers:

A good general reference about the psychology of pitch perception is the book:

B.C.J. Moore, An Introduction to the Psychology of Hearing, Academic Press, London, 1997.

This book is available in paperback and makes a good desk reference.

An algorithm implementation that matches a large body of psychoacoustical work, but which is computationally very intensive, is presented in the paper:

Malcolm Slaney and Richard Lyon, "A Perceptual Pitch Detector," Proceedings of the International Conference of Acoustics, Speech, and Signal Processing, 1990, Albuquerque, New Mexico. Available for ftp at Malcolm Slaney and Richard Lyon, "A Perceptual Pitch Detector," Proceedings of the International Conference of Acoustics, Speech, and Signal Processing, 1990, Albuquerque, New Mexico. Available for ftp at ftp://worldserver.com/pub/malcolm/ICASSP90.psc.Z

The definitive papers describing the use of such a perceptual pitch detector as applied to the classical pitch literature is in:

Ray Meddis and M. J. Hewitt. "Virtual pitch and phase sensitivity of a computer model of the auditory periphery. " Journal of the Acoustical Society of America 89 (6 1991): 2866-2682. and 2883-2894.

The current work that argues for a pure spectral method starts with the work of Goldstein:

J. Goldstein, "An optimum processor theory for the central formation of the pitch of complex tones," Journal of the Acoustical Society of America 54, 1496-1516, 1973.

Two approaches are worth considering if something approximating pitch is appropriate. The people at IRCAM have proposed a harmonic analysis approach that can be implemented on a DSP:

Boris Doval and Xavier Rodet, "Estimation of Fundamental Frequency of Musical Sound Signals," Proceedings of the 1991 International Conference on Acoustics, Speech, and Signal Processing, Toronto, Volume 5, pp. 3657-3660.

The classic paper for time domain (peak picking) pitch algorithms is:

B. Gold and L. Rabiner, "Parallel processing techniques for estimating pitch periods of speech in the time domain," Journal of the Acoustical Society of America, 46, pp 441-448, 1969.

Finally, a word of caution:

Pitch is not single-valued. We can hear a sound and match it to several different pitches. Imagine the number of instruments in an orchestra, each with its own pitch. Even a single sound can have more than one pitch. See for example Demonstration 27 from the ASA Auditory Demonstrations CD.