This thesis is an investigation of robust vector quantization, with the purpose of providing a system for the application of data compression and speech enhancement. Vector quantization is widely used in data compression systems. However, the performance of these systems will degrade in a noisy environment. The proposed robust vector quantization system solves the problem of optimal quantizatio...

We present a new Trellis Coded Vector Residual Quantizer (TCVRQ) that combines trellis coding and vector residual quantization. We introduce new methods for computing quantization levels and experimentally analyze the performances of our TCVRQ in the case of speech coding at very low bit rates. The results obtained show that transparent quantization of Linear Prediction (LP) parameters can be p...

In this paper we propose a new observations selection technique in Vector Quantization context. The main idea is to select observations that are not representatives of their classes. The cells generated by the Vector Quantization are divided in regions of rejection. A number of controlled experiments were performed demonstrating the proposed methodology potentiality. Key-Words: Observations Sel...

 In this paper, utilization of clustering algorithms for data fusion in decision level is proposed. The results of automatic isolated word recognition, which are derived from speech spectrograph and Linear Predictive Coding (LPC) analysis, are combined with each other by using fuzzy clustering algorithms, especially fuzzy k-means and fuzzy vector quantization. Experimental results show that the...

Storaging or transmission of speech signals at very low bit rate is a hot area in the field of speech processing. We used stochastic inter-frame interpolators and vector quantization (VQ) as a new method for developing a high quality 1200 BPS speech vocoder. The objective and subjecgtive test results show that performance of the new vocoder is compairable with 4800 BPS standard vocoders (as CELP).

This paper proposes novel lossy image compression technique using hybrid wavelet transform and vector quantization. First hybrid wavelet transform consisting of two different component transforms is generated and applied on color images. Discrete Kekre transform (DKT) and Discrete Cosine transform (DCT) play role of base and local transform respectively in hybrid wavelet transform. In transform...

This paper investigates the use of a new hybrid vector quantizer called switched multistage vector quantization (SWMSVQ) technique using hard and soft decision schemes, for coding of narrow band speech signals. This technique is a hybrid of switch vector quantization technique and multistage vector quantization technique. SWMSVQ quantizes the linear predictive coefficients (LPC) in terms of the...

We propose an extension of the Conn-Index to evaluate fuzzy cluster solutions obtained from fuzzy prototype vector quantization, whereas the original Conn-Index was designed for crisp vector quantization models. The fuzzy index explicitly takes the fuzzy assignments resulting from fuzzy vector quantization into account. This avoids the information loss which would occur if the original crisp in...

Storaging or transmission of speech signals at very low bit rate is a hot area in the field of speech processing. We used stochastic inter-frame interpolators and vector quantization (VQ) as a new method for developing a high quality 1200 BPS speech vocoder. The objective and subjecgtive test results show that performance of the new vocoder is compairable with 4800 BPS standard vocoders (as CELP).