scispace - formally typeset
Search or ask a question
Topic

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.


Papers
More filters
Proceedings ArticleDOI
25 May 2003
TL;DR: A novel LP model based coding technique is presented where the advantages of multiple nonorthogonal domain representations of LP coefficients and the prediction residuals are exploited in conjunction with vector quantization to yield considerable LP coding enhancement.
Abstract: Recently, signal processing in multiple nonorthogonal domains has been reported that further enhances the efficiency of waveform and Linear Prediction (LP) model based signal representation. In this contribution, a novel LP model based coding technique is presented where the advantages of multiple nonorthogonal domain representations of LP coefficients and the prediction residuals are exploited in conjunction with vector quantization to yield considerable LP coding enhancement. The proposed signal coding technique is applied to speech signals and the resulting performance improvement is clearly demonstrated in terms of the reconstruction quality for the same bit rate compared to the existing single domain vector quantization techniques.
Patent
30 Jan 2009
TL;DR: In this paper, a Code-Excited Linear Prediction CELP was used for decoding a voice code including a linear prediction parameter code, an adaptation code, and a gain code.
Abstract: PROBLEM TO BE SOLVED: To reproduce voice of high quality by a small information quantity in voice decoding for decoding a voice code including a linear prediction parameter code, an adaptation code and a gain code by a Code-Excited Linear Prediction CELP. SOLUTION: In decoding of the voice code by the Code-Excited Linear Prediction CELP, a noise degree on the voice code is evaluated in a decoding section based on the adaptation code, and different excited code books 22 and 23 are used in response to an evaluation result. COPYRIGHT: (C)2009,JPO&INPIT
Proceedings ArticleDOI
01 Nov 1993
TL;DR: In this article, the authors make objective and informal subjective measurements of the floating-point 16 kb/s LD-CELP performance with different bit-error rates and examine the sensitivity of this coder to errors in gain and shape codeword index parts and the effect of an error to future outputs of the coder.
Abstract: We make objective and informal subjective measurements of the floating-point 16 kb/s LD-CELP performance with different bit-error rates. We also study the sensitivity of this coder to errors in gain and shape codeword index parts and the effect of an error to future outputs of the coder. Based on the transmitted code statistics we attempt to minimize the annoying effects of codeword errors. Finally, we examine audible oscillations and distortions in the synthesized speech signal, again in the context of channel errors. >
Proceedings ArticleDOI
03 Jun 2015
TL;DR: Voice-excited LPC is the technique proposed in this paper that results in a low bit rate and a better signal to noise ratio and also provides accurate estimation of speech parameters and is computationally effective.
Abstract: Speech signal is the unique and special signal in communication system so it must be analyzed in order to extract its important parameters and to compress it for maximum utilization of available bandwidth. For this, there are various kinds of speech analysis and synthesis techniques that have been effectively used. Among all these techniques, Linear Predictive Coding (LPC) is the most powerful one to represent the speech signal at reduced bit rates while preserving the quality of the signal and also provides accurate estimation of speech parameters and is computationally effective. Voice-excited LPC is the technique proposed in this paper. This technique has been implemented using both male and female voices and trade-offs between bit rates, delay, power signal to noise ratio and complexity are analyzed. It results in a low bit rate and a better signal to noise ratio.
Patent
08 Dec 2006
TL;DR: An apparatus for encoding a voice and a method thereof are provided to have no need to attempt other methods for expanding a narrow band to a wide band or an audio band in a codec having necessity for fine bit rate extension, thereby acquiring algorithm consistency and reducing calculation amount reduction.
Abstract: An apparatus for encoding a voice and a method thereof are provided to have no need to attempt other methods for expanding a narrow band to a wide band or an audio band in a codec having necessity for fine bit rate extension, thereby acquiring algorithm consistency and reducing calculation amount reduction. A band division unit(100) divides an input signal into a high band signal and a low band signal. A narrow band encoding unit(105) encodes the low band signal using a narrow band voice codec based on CELP(Code Excited Linear Prediction). A frequency feature collecting unit(110) converts the high band signal into a frequency domain and calculates MDCT(Modified Discrete Cosine Transform) coefficients. Subband determination units(115,120) determine a subband for shape quantization based on the MDCT coefficients, and determine a subband for gain quantization based on the subband for the determined shape quantization. A gain quantization unit(125) performs gain quantization for the subband for the gain quantization. A bit allocation unit(130) allocates bits to the subband for the gain quantization according to the size of the gain quantization. A shape quantization unit(135) performs shape quantization for the subband for the shape quantization in an algebraical solution.

Network Information
Related Topics (5)
Decoding methods
65.7K papers, 900K citations
83% related
Data compression
43.6K papers, 756.5K citations
83% related
Signal processing
73.4K papers, 983.5K citations
83% related
Feature vector
48.8K papers, 954.4K citations
80% related
Feature extraction
111.8K papers, 2.1M citations
79% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
20226
20213
20207
201915
201810
201713