The Residual Encoder--An Improved ADPCM System for Speech Digitization

Cohn, David L.; Melsa, James L.

doi:10.1109/tcom.1975.1092919

Cited by 48 publications

(18 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The compressed speech samples are coded using an Adaptive Residual Coder [3]. The output of the ARC is a string of quantizer outputs q(k).…”

Section: System Configurationmentioning

confidence: 99%

“…Pat least one error per frame} = 1-(1-)" (3) Psingle error per frame} = n(6)(1_)r-1 (4) If a single error in the frame is corrected, the residual error becomes [1 _(1_c)n -nc(1-)fl-l].…”

Section: Transmission Errorsmentioning

confidence: 99%

“…The technique, which has been reported in a previous paper [1], is based on a combination of Time Domain Harmonic Scaling (TDHS) [2] and Adaptive Residual Coder (ARC) [3].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Mediumband speech encoding using time-domain harmonic scaling and adaptive residual coding for noisy channels

Melsa

Pande²

ICASSP '82. IEEE International Conference on Acoustics, Speech, and Signal Processing

Self Cite

View full text Add to dashboard Cite

This paper describes the study of an approach to speech digitization at mediumband bit rates of 9.6 kb/s to 16 kb/s for the noisy channels.The technique is based on a combination of Time Domain Harmonic Scaling and an Adaptive Residual Coding. The use of fixed wordsize codewords and the absence of any side information transmission makes the system structure quite simple.Simulation studies have shown that this technique produces communication quality speech at 9.6 kb/s and excellent quality speech at 16 kb/s with a channel bit-error-rate as high as 1%.

show abstract

“…The compressed speech samples are coded using an Adaptive Residual Coder [3]. The output of the ARC is a string of quantizer outputs q(k).…”

Section: System Configurationmentioning

confidence: 99%

“…Pat least one error per frame} = 1-(1-)" (3) Psingle error per frame} = n(6)(1_)r-1 (4) If a single error in the frame is corrected, the residual error becomes [1 _(1_c)n -nc(1-)fl-l].…”

Section: Transmission Errorsmentioning

confidence: 99%

See 1 more Smart Citation

Mediumband speech encoding using time-domain harmonic scaling and adaptive residual coding for noisy channels

Melsa

Pande²

ICASSP '82. IEEE International Conference on Acoustics, Speech, and Signal Processing

Self Cite

View full text Add to dashboard Cite

show abstract

“…If the bit rate is kept the same, the number of bits allowed per sample is doubled, and the performance of the coder can be improved significantly. The ARC structure was developed by Cohn and Melsa [3] and implemented in hardware by CODEX [4]. This structure involves adaptive linear predictor, pitch compensating adaptive quantizer [5] and adaptive source coding.…”

Section: Introductionmentioning

confidence: 99%

“…This is more closely related to speech quality measure than simple SNR.where y(m) is the compressed output and h(i:N0) is the window function, Equation(2)can be rewitten to indicate that only one multiplication and two additions are required per output sample. As long as the window function satisfies the TOES-ARC System Configuration ARC is a highly developed AOPCM system first copoeed by Cohn & Ifelsa[3] and implemented in hardware by CODEX. The underlying design principles of the ARC system is that only backward adaptation is used.…”

mentioning

confidence: 99%

Mediumband speech encoding using time domain harmonic scaling and adaptive residual coding

Melsa

Pande²

ICASSP '81. IEEE International Conference on Acoustics, Speech, and Signal Processing

Self Cite

View full text Add to dashboard Cite

show abstract

Pitch-Adaptive DPCM Coding of Speech With Two-Bit Quantization and Fixed Spectrum Prediction

Jayant¹

1977

Bell System Technical Journal

View full text Add to dashboard Cite

This paper is concerned with the utilization of speech waveform periodicities in differential pulse code modulation (dpcm) coding with 2‐bit adaptive quantization and time‐invariant spectrum prediction. Our work is based on computer simulations of dpcm codes. We have studied pitch detectors based on autocorrelation and an average magnitude difference function (amdf), and we have measured the benefits of predicting from a previous pitch period as functions of pitch‐period‐updating frequency and periodicity‐indicating thresholds (for autocorrelation and the amdf). We have compared several alternative methods of utilizing past quantized samples (in the present and previous pitch periods) for providing speech sample predictions. We find the following combination to be attractive for waveform coding at bit rates in the neighborhood of 16 kb/s: 2‐bit adaptive quantization with a one‐word (2‐bit dpcm word) memory, pitch detection performed on unquantized speech (preferably with an amdf criterion) and a prediction scheme that uses fixed three‐tap (short‐term) prediction for nonperiodic waveform segments, but switches to an appropriate one‐tap (long‐term) predictor upon the detection of strong periodicity. With four sample utterances, the latter procedure results in an average SNR (signal‐to‐noise ratio) gain of 3.75 dB over a non‐pitch‐adaptive encoder.

show abstract

The Residual Encoder--An Improved ADPCM System for Speech Digitization

Cited by 48 publications

References 0 publications

Mediumband speech encoding using time-domain harmonic scaling and adaptive residual coding for noisy channels

Mediumband speech encoding using time-domain harmonic scaling and adaptive residual coding for noisy channels

Mediumband speech encoding using time domain harmonic scaling and adaptive residual coding

Pitch-Adaptive DPCM Coding of Speech With Two-Bit Quantization and Fixed Spectrum Prediction

Contact Info

Product

Resources

About