Faculty of Engineering
Christian-Albrechts-University, Kiel
Institute for Circuit and System Theory
Logo der Technischen Fakultät
Description of the Lectures

Digital Speech-Signal Processing Zum Seitenanfang

The spacer
Lecturer(s) Prof. Dr.-Ing. U. Heute and research assistants.
Target group Students of Electrical Engineering and Computer Science, graduate level, also other students with basic knowledge in signals and systems theory.
Prerequisites "Vordiplom", lectures and exercises in "Signals and System" and "System theory"
Contents Spectral and statistical speech characteristics; fundamentals of physiology. Waveform-coding: linear PCM, improvements by nonlinearity, adaptive quantization, optimal quantization, vector quantization. Linear prediction and linear-predictive time-domain speech coding; analysis-by-synthesis, quantization of the residual signal by MPE, RPE; CELP-algorithm, gain-shape vector quantization, VSELP, ACELP. Frequency domain coding. Vocoder-techniques. Application of speech coding in mobile radio systems; bad-frame handling to improve the quality in case of corrupted transmission. Speech enhancement by noise reduction and echo cancellation.
  • Rabiner, L. R.; Schafer, R. W.: Digital Processing of Speech Signals; Prentice-Hall, 1978
  • Kondoz, A. M.: Digital Speech-Coding for Low Bit Rate Communications System; John Wiley & Sons, 1994

Zum Seitenanfang

The spacer