Code-excited linear prediction

About: Code-excited linear prediction is a research topic. Over the lifetime, 2025 publications have been published within this topic receiving 28633 citations. The topic is also known as: CELP.

Vector quantization codebook generation method

Abstract: A CELP speech coder or decoder comprising an adaptive codebook; a random codebook; a synthesis filter for receiving a random code vector generated from the random codebook so as to perform LPC synthesis; wherein said random codebook is formed by an excitation vector generator comprising fixed waveform arranging means for arranging one or more fixed waveforms (v1,v2,v3) stored in a fixed waveform storage means in accordance with the positions (P1,P2,P3) and polarities of an input impulse vector. The CELP speech coder/decoder further comprising means for determining a voiced/unvoiced characteristic of the input speech and that the random codebook generates the random code vector in accordance with the determined voiced/unvoiced characteristic.

Artificial Bandwidth Extension Based Packet Loss Concealment for CELP-Type Speech Coders

Abstract: In this paper, a packet loss concealment (PLC) algorithm is proposed to improve the quality of decoded speech when packet losses occur in a code- excited linear prediction (CELP)-type speech coder. The proposed PLC algorithm is based on artificial bandwidth extension (ABE) from narrowband to wideband, consisting of packet loss concealment in the narrow-band, ABE in the modified discrete cosine transform (MDCT) domain, and smoothing of wideband MDCT coefficients with those of the last good frame. The effectiveness of the proposed PLC algorithm is demonstrated by an informal listening test.

A Wide Band Speech Coding Technique using Low Delay Code Excited Linear Predictive Algorithm (LD-CELP).

Abstract: A fair level of speech quality is desired in speech transmission for mobile voice services. The effective utilization of bandwidth and higher bit rate is must for a best quality speech coder. But at a time the both requirements are not fulfilled in desired format. The research is ongoing in the area of designing speech coder’s. In general the CELP is an algorithm to design a good quality speech coder. From 80’s to present the advancement in this technique is going on. In this paper a wide band speech coding technique is proposed using LD-CELP algorithm. The overall performance of LD-CELP (16Kbps) is summarized and computed on MATLAB version R2016a with parameters MSE and SNR. In conclusion we observe that SNR for LD-CELP is not much better and enhancement in this is necessary.

Pole-zero analysis of speech by linear predictive coding.

Abstract: A method for generating pole-zero models for speech segments is described based on extracting Linear Predictive Coding (LPC) coefficients. Cepstral coefficient are generated from the LPCs and are used to obtain the negative derivative of the phase, thus enabling the spectrum to be split into a pole part and a zero part. Word classification by pole-zero tables, and their use for word recognition are described.

Decoding apparatus and method

Abstract: In a speech codec, the total number of transmitted bits is to be reduced to decrease the average amount of bit transmission by imparting a relatively large number of bits to the voiced speech having a crucial meaning in a speech interval and by sequentially decreasing the number of bits allocated to the unvoiced sound and to the background noise. To this end, the present invention provides a decoding apparatus for decoding encoded bits with different bit allocation to parameters of an unvoice interval and parameters of a voiced interval, including verifying means for verifying whether an interval in said encoded bits is a speech interval or a background noise interval and decoding means for decoding the encoded bits at the background noise interval by using LPC coefficients received at present or at present and in the past, CELP gain indexes received at present or at present and in the past and CELP shape indexes generated internally at random if the information indicating the background noise interval is taken out by said verifying means.