This correspondence presents quantization scheme for encoding line spectral parameters used in linear predictive coding (LPC) of speech. The scheme is based on low-dimensionality regular-point lattices. The algebraic ...
详细信息
This correspondence presents quantization scheme for encoding line spectral parameters used in linear predictive coding (LPC) of speech. The scheme is based on low-dimensionality regular-point lattices. The algebraic codebook need not be stored, and the optimum codevector is found through simple rounding of the input vector. Thus, the scheme results in significant savings of memory and reduced computational complexity when compared to traditional vector-quantizer solutions. The quantizer achieves an average spectral distortion of about 1 dB at 28 b/frame for the telephone bandwidth.
A systematic and unified approach which accomplishes performance monitoring, performance improvement and fault prediction in control systems is proposed. The feature vector which is a vector formed of the coefficients...
详细信息
A systematic and unified approach which accomplishes performance monitoring, performance improvement and fault prediction in control systems is proposed. The feature vector which is a vector formed of the coefficients of the estimate of the sensitivity function and the influence matrix which is the Jacobian of the feature vector with respect to the physical parameter are shown to contain the relevant information to realize an autonomous control system. The feature vector is estimated using a robust, accurate and reliable linear predictive coding Algorithm (LPCA). The influence matrix is computed by perturbing the physical parameters one at a time and estimating the feature vectors for each case. The proposed scheme is evaluated both on simulated as well as on actual control systems.
This paper describes a high-quality 8-kb/s speech coder called conjugate structure code-excited linear prediction (CS-CELP) with 10-ms frame length, To provide a short delay and high quality under both error-free add ...
详细信息
This paper describes a high-quality 8-kb/s speech coder called conjugate structure code-excited linear prediction (CS-CELP) with 10-ms frame length, To provide a short delay and high quality under both error-free add channel error conditions, it uses three dew schemes: line spectrum pair (LSP) quantization using interframe prediction, preselection in the codebook search, and gain vector quantization (VQ) with backward prediction;LSP parameters are quantized by using multistage VQ with moving-average (MA) prediction, This scheme can operate efficiently with various frequency responses of speech. The preselection of the codebook reduces computational complexity and improves robustness to channel errors. The gain VQ with backward prediction can provide high quality and robustness without transmission of input speech power information, A conjugate structure for both random codebook and gain codebook is introduced to improve the ability to handle random bit errors and to reduce codebook storage memory requirements, Subjective testing indicates that the quality of this coder is equivalent to that of 32-kb/s adaptive differential pulse code modulation (ADPCM) under error-free conditions, Testing has further demonstrated that the coder is robust against random bit errors.
This paper addresses the techniques used to produce a relatively low bit rate for speech coding. The technique initially explored in this work is the linearpredictive coder, the most popular method referenced to date...
详细信息
This paper addresses the techniques used to produce a relatively low bit rate for speech coding. The technique initially explored in this work is the linearpredictive coder, the most popular method referenced to date. This technique uses algorthims that describe the speech production process during voice and unvoiced sounds. The linearpredictive coder attempts to approximate the vocal tract filter over a short period of time. The model, coupled with a specific excitation, can be used for speech synthesis.
This paper introduces a new parametric formulation to linear predictive coding (LPC). Instead of the usual linear prediction filter coefficients, a new scheme similar to the line spectral frequencies (LSF), that is in...
详细信息
This paper introduces a new parametric formulation to linear predictive coding (LPC). Instead of the usual linear prediction filter coefficients, a new scheme similar to the line spectral frequencies (LSF), that is insensitive to quantization is proposed. However, in the present case, unlike the LSF scheme, an additional positive weighting factor gives extra freedom for coding. This new representation for LPC modeling is shown to be always stable under quantization.
Recent work has shown that multi-band excitation (MBE) is capable of synthesizing high quality speech in the range of 4.8-8.0 kbps. In this paper, the multi-band excitation and linear predictive coding (MBE-LPC) model...
详细信息
Recent work has shown that multi-band excitation (MBE) is capable of synthesizing high quality speech in the range of 4.8-8.0 kbps. In this paper, the multi-band excitation and linear predictive coding (MBE-LPC) model, which used the LPC analysis to obtain the prediction residual, and uses the MBE model to estimate the residual spectrum is presented. Our motivation is to improve the excitation model of LPC vocoders in the frequency domain. Based on the MBE-LPC model, a 5.4 kbps speech coding is presented. An adaptive postfilter is used to improve the perceptual quality of the decoded speech. Informal listening tests show that the perceptual quality of the decoded speech of the proposed coder is better than that of the 4.15 kbps improved (IMBE) coder.< >
In this correspondence, a new method of analysis of speech is proposed that will bring out variations in vocal tract system characteristics in short (2-4 ms) segments. In this method, the source and system components ...
详细信息
In this correspondence, a new method of analysis of speech is proposed that will bring out variations in vocal tract system characteristics in short (2-4 ms) segments. In this method, the source and system components of the speech signal are suitably windowed to reduce the effects of truncation of conventional waveform windowing.
The authors describe an improved pitch detection algorithm for efficient multiband excitation (MBE) coding of speech. The improved algorithm adds a corrective measure to the error measure for spectrum matching employe...
详细信息
The authors describe an improved pitch detection algorithm for efficient multiband excitation (MBE) coding of speech. The improved algorithm adds a corrective measure to the error measure for spectrum matching employed in conventional MBE pitch analysis. This corrective measure effectively applies equal weighting Lo all harmonic bands by normalising the error energy in each band. The result is the reduction of gross pitch errors owing to pitch doublings,An additional advantage of using this corrective measure is that, because this measure is based on a sum-of-product formula, comparisons of matching scores may be performed with partial-sums computed during the evaluation of the measure, therefore, facilitating fast searching of the optimum pitch period, Simulation results show that, with the deployment of the improved algorithm, pitch tracking procedure is no longer needed;and the coding delay of the MBE coder;is significantly reduced.
The search for better and more robust performance of speech recognition systems is ongoing, Much of the improvement is likely to come from better acoustic feature analysis, In this letter, the results from a significa...
详细信息
The search for better and more robust performance of speech recognition systems is ongoing, Much of the improvement is likely to come from better acoustic feature analysis, In this letter, the results from a significant experiment are reported;these show how a warped-DFT analysis outperforms an LPC-cepstral analysis in a significant way, supporting results by other researchers for different recognition tasks, An analysis of nasal-letter performance is used to show the development of the warped-DFT feature analysis.
The authors present a new secondary pulse excitation for linear prediction based analysis by synthesis speech coders. The structure of the excitation has been specifically designed to model characteristics in the spee...
详细信息
The authors present a new secondary pulse excitation for linear prediction based analysis by synthesis speech coders. The structure of the excitation has been specifically designed to model characteristics in the speech waveform which the LTP memory fails to adequately represent. This is achieved using an excitation vector simply consisting of two pulses.
暂无评论