A 2.4 kbps high-quality speech coder

Abstract
An algorithm for coding speech at 2.4 kbps is presented. By combining in a new way well-known coding strategies from both medium and low bit rate speech coding, a very promising configuration is developed. The coder is fundamentally a baseband coder where short-term correlation is predicted by LPC (linear predictive coding) analysis. Coding of the short-term parameters is performed by vector quantization. An open-loop long-term predictor is applied to the lowpass filtered short-term residual to reduce the quasi-periodic pitch structure before the signal is down-sampled. A novel method for coding the down-sampled residual based on voiced/unvoiced classification is proposed. For unvoiced frames a simple white Gaussian codebook is applied, and for voiced frames a new pulse codebook has been designed. The coding algorithm is demonstrated to be robust and to produce high speech quality at 2.4 kbps.

This publication has 3 references indexed in Scilit: