语音线性预测编码课件.pptVIP

Speech Signal Processing—Lecture 10 Linear Predictive Coding (LPC)-Introduction LPC Methods LPC methods are the most widely used in speech coding, speech synthesis, speech recognition, speaker recognition and verification and for speech storage LPC methods provide extremely accurate estimates of speech parameters, and does it extremely efficiently basic idea of Linear Prediction: current speech sample can be closely approximated as a linear combination of past samples, i.e., LPC Methods Basic Principles of LP LP Basic Equations LP Estimation Issues Solution for {αk} Solution for {αk} Solution for {αk} Solution for {αk} Autocorrelation Method Autocorrelation Method Autocorrelation Method Covariance Method Covariance Method Cholesky Decomposition Example Cholesky Decomposition Cholesky Decomposition Example Covariance Method Minimum Error Solution to Autocorrelation Method Autocorrelation Example Comparisons Between LP Methods the various LP solution techniques can be compared in a number of ways, including the following: computational issues numerical issues stability of solution number of poles (order of predictor) window/section length for analysis LP Solution Computations LP Solution Comparisons stability guaranteed for autocorrelation method cannot be guaranteed for covariance method; as window size gets larger, this almost always makes the system stable choice of LP analysis parameters need 2 poles (for 1 kHz) for each vocal tract resonance below Fs/2 need 3-4 poles to represent source shape and radiation load use values of p ≈ 13-14 Prediction Error Signal Behavior LP Analysis of Speech Frequency Domain Interpretation Frequency Domain Interpretation Frequency Domain Interpretation Comparisons of Spectral Analysis Methods LPC Synthesis Formant Analysis Using LPC * H(z) assume N1≈ N2p; choose values of N1=300, N2=300, p=10 computation for covariance method ≈ N1p+p3 ≈ 4000 *,+ autocorrelation method ≈ N2p+p2 ≈ 3100 *,+ note error signal near time=0 for aut