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
Book ChapterDOI
23 Nov 2016
TL;DR: A Forward Error Correction (FEC)-based technique which relies on energy constraint to determine frame onset which will be considered for sending the FEC information and greatly improves the CELP-based codec robustness to packet losses with no increase in coder storage capacity.
Abstract: The strong interframe dependency present in Code Excited Linear Prediction (CELP) codecs renders the decoder very vulnerable when the Adaptive Codebook (ACB) is desynchronized. Hence, errors affect not only the concealed frame but also all the subsequent frames. In this paper, we have developed a Forward Error Correction (FEC)-based technique which relies on energy constraint to determine frame onset which will be considered for sending the FEC information. The extra information contains an optimized FEC pulse excitation which models the contribution of the ACB to offer a resynchronization procedure at the decoder. In fact, under the energy constraint the number of Fixed Codebook (FCB) pulses can be reduced in order to be exploited by the FEC intervention. In return, the error propagation is considerably prevented with no overload of added-pulses. Furthermore, the proposed method greatly improves the CELP-based codec robustness to packet losses with no increase in coder storage capacity.
Proceedings Article
01 Jan 2011
TL;DR: A new wideband speech coding model to efficiently compress acoustic inventories for concatenative unit selection text-to-speech (TTS) synthesis system and a hybrid coding framework is also suggested.
Abstract: This paper suggests a new wideband speech coding model to efficiently compress acoustic inventories for concatenative unit selection text-to-speech (TTS) synthesis system. To fulfill the requirements of TTS synthesizer such as partial segment decoding and random access capability, a non-predictive scheme was adopted which combines the adaptive Quasi-Harmonic Model (aQHM) with the innovative codebook (ICB) model. aQHM plays a major role in modeling pitch harmonic components, and ICB compensates, in a closed-loop way, for the modeling error of aQHM. This is especially important in transient or unvoiced regions. To further improve the coding efficiency, a hybrid coding framework is also suggested. Results from a large French speech database show that the proposed algorithm provides similar speech quality to the high quality AMR-WB codec while it supports the random access capability.
01 Jan 2014
TL;DR: In this paper, the C program of the coder was compiled using the GCC compiler in the Linux environment and executed using the GNU Make utility, and the bit rate obtained is 16 kb/s, i.e. a reduction by 4 times of the original bitrate.
Abstract: used to extract details regarding the quality of the sound synthesized by the coder. The C program of the coder was compiled using the GCC compiler in the Linux environment and executed using the GNU Make utility. File compression ratio of 5 was achieved. The bit rate obtained is 16 kb/s, i.e. a reduction by 4 times of the original bitrate. The perceptual quality of the reconstructed audio output is very good, and this is evident from the spectrogram, pitch, intensity and formant waveforms obtained.
Patent
14 Apr 2000
TL;DR: In this paper, a speech code series is decoded to a pitch period, a LSP coefficient, code numerals or the like by the synthesizing part 115 to reproduce the speech sound in the CELP system.
Abstract: The reception part 106 receives a code series that has propagated on the communication path to be transmitted to the separation part 107. The separation part 107 separates the code series into a speech code series and text information to be outputted to the synthesizing part 115 and the language analysis part 109, respectively. The speech code series is decoded to a pitch period, a LSP coefficient, code numerals or the like by the synthesizing part 115 to reproduce the speech sound in the CELP system. On the other hand, the text information is converted into information of pronunciation and accent by the language analysis part 108, which is added to prosody information such as phoneme time length and pitch pattern by the prosody generation part 110. The LSP coefficient, code numerals or the like which are suitable for the phoneme, are read out from the segment DB 114 by the segment read-out part 113, and the pitch frequency is taken out from the prosody information to be inputted to the synthesizing part 115 so as to be synthesized into a speech sound.
Proceedings ArticleDOI
05 May 1996
TL;DR: The highly integrated JADE (Joint Audio Decoder-Encoder) device implements audio compression using either G. 711, G.722 or G.728 compression algorithms, opening up the market for low cost digital telecommunications applications.
Abstract: The highly integrated JADE (Joint Audio Decoder-Encoder) device implements audio compression using either G.711, G.722 or G.728 compression algorithms. The integration of peripherals, interfaces and memories with a fixed point DSP (digital signal processor) to implement complex speech algorithms such as G.728, opens up the market for low cost digital telecommunications applications. Using the G.728 Low Delay Code Excited Linear Prediction (LD-CELP) algorithm, the device compresses/decompresses a digitized PCM (64 kbps) or linear (128 kbps) voice signal to/from a 16 kbps bit stream. The device has been fabricated in a 0.5 /spl mu/m n-well CMOS process.

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