The Intel IPP functions described in this section implement building blocks that can be used to create speech codecs compliant with the ITU-T Recommendation G.729 (see [ITU729], [ITU729A], [ITU729B]).
These functions are used in the Intel® IPP G.729 Speech Encoder-Decoder sample downloadable from http://www.intel.com/cd/software/products/asmo-na/eng/220046.htm.
The list of these functions is given in the table below.
Function Base Name | Operation |
---|---|
Basic Functions | |
DotProd_G729 | Computes the dot product of two vectors. |
Interpolate_G729 | Computes the weighted sum of two vectors. |
Linear Prediction Analysis Functions | |
AutoCorr_G729 | Estimates the autocorrelation of a vector. |
LevinsonDurbin_G729 | Calculates LP coefficients from the autocorrelation coefficients. |
LPCToLSP_G729 | Converts LP coefficients to LSP coefficients. |
LSFToLSP_G729 | Converts Line Spectral Frequencies to LSP coefficients. |
LSFQuant_G729 | Performs quantization of LSF coefficients. |
LSFDecode_G729 | Decodes quantized LSFs. |
LSFDecodeErased_G729 | Reconstructs quantized LSFs in case when a frame is erased. |
LSPToLPC_G729 | Converts LSP coefficients to LP coefficients. |
LSPQuant_G729 | Quantizes the LSP coefficients. |
LSPToLSF_G729 | Converts LSP coefficients to LSF coefficients. |
LagWindow_G729 | Applies 60Hz bandwidth expansion. |
Codebook Search Functions | |
OpenLoopPitchSearch_G729 | Searches for an optimal pitch value. |
AdaptiveCodebookSearch_G729 | Searches for the integer delay and the fraction delay, and computes the adaptive vector. |
DecodeAdaptiveVector_G729 | Restores the adaptive codebook vector by interpolating the past excitation. |
FixedCodebookSearch_G729 | Searches for the fixed codebook vector. |
GainCodebookSearch_G729 | Finds pitch and fixed gains. |
ToeplizMatrix_G729 | Calculates 616 elements of the Toepliz matrix for the fixed codebook search. |
Codebook Gain Functions | |
DecodeGain_G729 | Decodes the adaptive and fixed-codebook gains. |
GainControl_G729 | Calculates adaptive gain control. |
GainQuant_G729 | Quantizes the codebook gain using a two-stage conjugate-structured codebook. |
AdaptiveCodebookContribution_G729 | Updates target vector for codebook search by subtracting adaptive codebook contribution. |
AdaptiveCodebookGain_G729 | Calculates the gain of the adaptive-codeblock vector and the filtered codebook vector. |
Filter Functions | |
ResidualFilter_G729 | Implements an inverse LP filter and obtains the residual signal. |
SynthesisFilter_G729 | Reconstructs the speech signal from LP coefficients and residuals. |
LongTermPostFilter_G729 | Restores the long-term information from the old speech signal. |
ShortTermPostFilter_G729 | Restores speech signal from the residuals. |
TiltCompensation_G729 | Compensates for the tilt in the short-term filter. |
HarmonicFilter | Calculates the harmonic filter. |
HighPassFilterSize_G729 | calculates the G729 high-pass filter size. |
HighPassFilterInit_G729 | Initializes the high-pass filter. |
HighPassFilter_G729 | Performs G729 high-pass filtering. |
IIR16s_G729 | Performs IIR filtering. |
PhaseDispersionGetStateSize_G729D | Queries the memory length of the phase dispersion filter. |
PhaseDispersionInit_G729D | Initializes the phase dispersion filter memory. |
PhaseDispersionUpdate_G729D | Updates the phase dispersion filter state. |
PhaseDispersion_G729D | Performs the phase dispersion filtering. |
Preemphasize_G729A | Computes pre-emphasis of a post filter. |
WinHybridGetStateSize_G729E | Queries the length of the hybrid windowing module memory. |
WinHybridInit_G729E | Initializes the hybrid windowing module memory. |
WinHybrid_G729E | Applies the hybrid window and computes autocorrelation coefficients. |
RandomNoiseExcitation_G729B | Initializes a random vector with a Gaussian distribution. |
FilteredExcitation_G729 | Computes filtered exitation. |
Copyright © 2000 - 2011, Intel Corporation. All rights reserved.