High-quality speech codec modules operating at 16 and 8 kb/s have been developed using an adaptive predictive coding with adaptive bit allocation (APC-AB) scheme. An optimized APC-AB algorithm is studied that reduces ...
详细信息
High-quality speech codec modules operating at 16 and 8 kb/s have been developed using an adaptive predictive coding with adaptive bit allocation (APC-AB) scheme. An optimized APC-AB algorithm is studied that reduces processing complexity while maintaining speech quality. The coding algorithm is implemented in two digital signal processors (DSPs). The DSP chips, a framing LSI circuit, a PCM codec, and some peripheral ICs are integrated in each of two compact packages, i.e. codec modules, operating at 16 or 8 kb/s. The codec module size is as small as 80 mm*50 mm*12 mm, and its typical power consumption is 500 mW using 2- mu m CMOS LSI technology. At 16 kb/s this APC-AB codec achieves high speech quality, close to that of a 7-bit mu -law PCM. The codec modules are expected to be used for various applications such as customer premises multiplexers for digital leased lines, digital mobile radio, and stored-and-forward-message systems (voice-mail systems).< >
An adaptive technique for image sequence coding that is based on vector quantization is described. Each frame in the sequence is first decomposed into a set of vectors. A codebook is generated using the vectors of the...
详细信息
An adaptive technique for image sequence coding that is based on vector quantization is described. Each frame in the sequence is first decomposed into a set of vectors. A codebook is generated using the vectors of the first frame as the training sequence, and a label map is created by quantizing the vectors. The vectors of the second frame are then used to generate a new codebook, starting with the first codebook as seeds. The updated codebook is then transmitted. At the same time, the label map is replenished by coding the position and the new values of the labels that have changed from one frame to the other. The process is repeated for subsequent frames. Experimental results for a test sequences demonstrate that the technique can track the changes and maintain a nearly constant distortion over the entire sequence.< >
A Code-Excited Linear predictive (CELP) coder is developed for transmission purposes and its connection with other predictive coding schemes as well as vector quantization is clarified. At a rate of one bit per sample...
详细信息
A Code-Excited Linear predictive (CELP) coder is developed for transmission purposes and its connection with other predictive coding schemes as well as vector quantization is clarified. At a rate of one bit per sample a codebook composed of 2 L waveforms of L samples is specified taking their values in {+1, −1}. Each waveform i deduced from the L bits index by a one-to-one correspondence between each bit of the index and each sample of the waveform. By construction the codebook is shown to have an inherent robustness against transmission errors and its internal algebraic structure leads to efficient and fast algorithms for selecting the optimum excitation. Both objective and subjective results confirm the high level of performances obtained by the 16 kbit/s CELP coder in different realistic transmission conditions as transmission with errors and ambient noise. Der Artikel beschreibt ein Verfahren zur prädiktiven Codierung mit Codebuch-Anregung (CELP) und erläutert seine Beziehung zu anderen Codierungsverfahren mit linearer Praädiktion sowie zur Vektorquantisierung. Für eine Rate von einem bit pro Abtastwert wird ein Codebach bestehend aus 2 L Repräsentationsvektoren mit L Komponenten aus {−1, + 1} spezifiziert. Jeder Wellenzug wird aus dem L bits wertigen Index durch eine eindeutige Abbildung zwischen jedem Bit des Index und jedem Abtastwert der Wellenform abgeleitet. Aufgrund dieses Zuordnungsprinzips ergibt sich ein Codebuch, das sich robust gegenüber Übertragungsfehlern verhält, und seine innere algebraische Struktur erlaubt die Entwicklung von sehr schnellen Algorithmen für die Wahl der optimalen Wellenform. Die objektiven und subjektiven Ergebnisse bestätigen das hohe Leistungsniveau des CELP-Coders mit einer Übertragungsrate von 16 kbit/s bei verschiedenen realistischen Übertragungsverhältnissen sowie bei Übertragungsfehlern und Umgebungsgeräuschen. Cet article décrit une structure de codeur prédictif excité par codes pour des applications en transmission et c
The compression and decompression of gray-level-image files in the spatial domain using three different methods is examined. The first method uses the average weights of neighboring pixels to calculate the value of th...
详细信息
The compression and decompression of gray-level-image files in the spatial domain using three different methods is examined. The first method uses the average weights of neighboring pixels to calculate the value of the current one. The second method combines weights and a delta value to estimate the missing pixel. The purpose of the delta value is to enhance the quality at the edges. The third method uses adaptive vector quantization. Codebooks of representative vectors are generated for different portions of the image. The performance of the coder is estimated in terms of signal-to-noise ratio. coding parameters such as vector dimension, number of representative vectors, and searching technique are discussed. Compression ratios are examined as a function of signal-to-noise ratio, and running time.< >
A speech-coding algorithm based on the introduction of a vector quantizer into an ADPCM (adaptive differential pulse code modulation) configuration is presented. This vector ADPCM (VADPCM) algorithm is directed toward...
详细信息
A speech-coding algorithm based on the introduction of a vector quantizer into an ADPCM (adaptive differential pulse code modulation) configuration is presented. This vector ADPCM (VADPCM) algorithm is directed toward low-complexity, low-delay (0-5 ms), 16-kb/s applications. An analysis-by-synthesis configuration is used to allow the vector quantizer to operate with the usual scalar linear predictor. Performance/complexity tradeoffs are described. Methods for reducing the implementation complexity to the level of the standard 32-kb/s CCITT algorithm are indicated. Noise feedback, postfiltering, and gain-adaptive vector quantization are used to improve the performance while maintaining low complexity.< >
Multichannel 2-D linear prediction and its application to predictive coding of color images are described. Two algorithms are compared. In the first the whole frame of the image is divided into subframes and predictor...
详细信息
Multichannel 2-D linear prediction and its application to predictive coding of color images are described. Two algorithms are compared. In the first the whole frame of the image is divided into subframes and predictor coefficients are computed separately for each. In the second, predictor coefficients are obtained for the whole frame of the image. Both methods have the disadvantage that the coefficients must be computed in real time and transmitted to the receiver as side information, which significantly increases the complexity of the coding system. An alternative is considered that uses a fixed set of prediction matrices, i.e. a set that does not depend on the specific image being coded. In this way both receiver and transmitter have linear prediction matrices and no side information has to be transmitted. Such prediction matrices can be generated by various averaging methods discussed. The results of this coding are compared to that resulting from the previous two methods.< >
The temporal and spatial redundancies of image sequences can be reduced by prediction and vector quantization, respectively. The robustness of the coder is increased through the use of a hybrid two-stage vector quanti...
详细信息
The temporal and spatial redundancies of image sequences can be reduced by prediction and vector quantization, respectively. The robustness of the coder is increased through the use of a hybrid two-stage vector quantizer. The first stage is of the Linde-Buzo-Gray type. The second stage encodes the residual error of the first stage using a lattice vector quantizer.
Three dimensional display of moving images greatly enhances realism and adds a unique sense of "presence". Three dimensional video systems have been kept from widespread application by two technical problems...
详细信息
Three dimensional display of moving images greatly enhances realism and adds a unique sense of "presence". Three dimensional video systems have been kept from widespread application by two technical problems, the need for glasses, viewing hoods, or other cumbersome devices for image steering, and the high bandwidths needed for transmission. Devices that avoid the discomfort of headgear by using autostereoscopic (pseudo-holographic) displays are known, but these methods require even higher bandwidths to be effective. This paper introduces the use of digital predictive coding as a means of data compression for the transmission or storage of a set of spatially related images needed for an autostereoscopic display. (Interframe coding without frame memories.) The algorithms, implementations, and application of a new sort of predictor called Disparity Corrected Prediction are described.
The reasons underlying the success of predictive coding are outlined. Proper attention to auditory perception and masking of the quantizing noise have resulted in coders with high quality at 0.5 bits/residual sample. ...
详细信息
The reasons underlying the success of predictive coding are outlined. Proper attention to auditory perception and masking of the quantizing noise have resulted in coders with high quality at 0.5 bits/residual sample. Ever more powerful chips have transported these basic principles into the realm of real-time reality. For sustained progress, current research should focus on 1. The effects of dynamic masking: during and immediately following rapid transitions in the speech spectrum the ear seems more tolerant to rough coding. The saving in average information rate could be substantial. 2. Better algebraically structured codes: codes are needed that are fast, simple to implement and allow the incorporation of subjective criteria. 3. Efficient coding of the parameters: because of the great success with coding the residual, sparse representation of the predictor coefficients is now paramount.
We have developed a new speech coder which significantly enhances Adaptive predictive coding (APC) by using vector quantization. The coder, called Vector APC (VAPC), gives very good speech quality at 9.6 kb/s and reas...
详细信息
We have developed a new speech coder which significantly enhances Adaptive predictive coding (APC) by using vector quantization. The coder, called Vector APC (VAPC), gives very good speech quality at 9.6 kb/s and reasonably good quality at 4.8 kb/s. In VAPC, redundancy is first removed by a long-delay predictor and then by a short-delay predictor; the prediction residual is then quantized by a gain-adaptive vector quantizer. In the receiver, decoded residual vectors are used to excite a synthesis filter to obtain the coded speech. The computations required by VAPC are only in the order of 2 to 4 million flops per second. Because of its low complexity and high speech quality, VAPC may offer a low-complexity alternative to Code-Excited Linear Prediction (CELP) at low bit rates.
暂无评论