A novel speech codec of high quality and low complexity——PS ACELP (pre search algebraic code excited linear prediction) is presented in this paper, which is suitable for software based compressing and replaying speech on H/PC.
729 speech codec (encoder and decoder) is an algorithm recommendation for the coding of speech signals at 8kbitls bit rate using Conjugate-Structure Algebraic-Code-Excited Linear-Predication (CS-ACELP), which was approved by 1W (International Telecommunication Union) Study Group 15 on the 19th of March 1996. 1W G. 729 has been widely used in the field of multimedia communication.
Recently~ there is the great interest in developing toll-quality speech coders at rates of 4kbps and below, and which are expected to be widely exploited in applications such as visual telephone~ mobile and personal communications.
Finally we give an evaluation of the real-time MCPS, memory requirements, processing delays and associate trade-offs of the coder using this DSP, and it is helpful for constructing other speech coders.
This paper introduces the principles of two variable-rate speech coders, AMR and SMV, in 3rd mobile communication system. AMR was compared with SMV in some aspects including complexity,error concealment of lost frames,system delay and synthesized speech quality.
The first stage of the structure consists of a core speech coder which provides a minimum output bit rate and acceptable performance on clean speech inputs.
The GSM speech coder for digital mobile telephones has been designed on a custom DSP using an environment for development of arbitrary processor architectures.
The contribution of this paper is in applying parameter replacement techniques to speech that is compressed by the Federal Standard 1016 CELP speech coder, protected by Reed-Solomon codes, and transmitted over a wireless channel.
Due to adopting these optimizing methods and programming skills, we have implemented a high-speed speech codec that can process concurrently 20 voice channels with single TMS320C6201 chip in IP telephony gateway.
This paper describes methods for mode selection in multirate speech codecs, such as the AMR (Adaptive Multi-Rate), that is the mandatory speech codec selected in 3GPP (3rd Generation Partnership Project) defined mobile networks.
Variable Rate (VR) speech coders are classified into: source-controlled VR coders where the rate is selected depending on the local character of the speech, and network-controlled VR coders where an external control signal selects the coding rate.
This paper gives a new digital speech coder with real time operation at 9.6Kb/s data rate.The system consists of time-domain harmonic scaling(TDHS)and sub-band coding (SBC)-two different speech compression algorithm.The basic principle of TDHS and several methods of tone detection are described.Finally,several modes of the system executed by DSP in real time are discussed in the paper.
Today, we are living with a informational society. Highly effective speech coding system isbecoming more and more important. Because of special characters of neural network (NN) such as parallel pro-cessing.distributed register and self-adapting capability etc, neural network has greatest potential in many area.Based on the pdf of speech parameters, a new neural network, Classification Neural Network-CNN, is presentedin the paper, and the quality for apeech coder makes improvement significantly.
The differences between multi-band excited(MBE)speech model and pitch excited linearpredictive coding(LPC)speech model,and the advantages of MBE model over the conven-tional excited model are presented in this paper.The principles and algorithms of speech analy-sis/synthesis based upon MBE model are studied.The approaches ofspeech coding at 4.8 kbpsare presented and simulated with computer. Experimental results show that the method ofMBE speech coding at4.8 kbps is reliable and its synthesized speech is hig...