On the predictor for the waveform coding of speech signals by using the dual first order difference values

Bae, Myung-Jin; Lee, M.; Min, Kwan-Sik; Ann, Sou Guil

doi:10.1109/tencon.1992.271986

Cited by 1 publication

(1 citation statement)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the synthesis, prediction error dA(n) coded in the transmitter is decoded at the receiver, and the result value adds the error e(n) generated in the decoder to the compensated prediction error d'(n) after quantizing. Namely, prediction error d*(n) ,after performing encode and decode, is In the predictor which uses the dual autocorrelation method [6], we predict the present sample by using the most adjacent samples. This method does not require convergence time which appears generally in the linear predictor.…”

Section: N)-{s(n-l)+s(n+l)}/2+{s(n+l)mentioning

confidence: 99%

On a new predictor for the waveform coding of speech signal by using the dual autocorrelation and the sigma-delta technique

Bae

Kim

Jeon

et al.

Proceedings of IEEE International Symposium on Circuits and Systems - ISCAS '94

View full text Add to dashboard Cite

The speech waveforms are highly correlated between the adjacent samples. One way of increasing the correlation in speech signals is to simply integrate the input signal prior to coding. The integrated values can be removed by conventional differentiation at the receiver. This emphasizes the low frequencies of speech signals and increases the correlation between adjacent samples. The above arrangement is called as a sigmadelta technique.In this paper, we propose a new predictor which uses such characteristics of dual autocorrelation and the sigmadelta technique. That is, we integrate input signals prior to coding ,and then predict the present intergrate sample by using two samples, one past and one next. The proposed predictor has higher mean prediction gain of 8.65dB than that of the CCI'IT-Recommendation ADPCM.

show abstract

Section: N)-{s(n-l)+s(n+l)}/2+{s(n+l)mentioning

confidence: 99%