DCT lifting, Why & How

Intel® Integrated Performance Primitives

Deliberate problems developing high-performance vision, signal, security, and storage applications.

DCT lifting, Why & How

293 Views

I am working on an application to extract MFCC features

Having in mind ippsDCTLifterInitAlloc_Mul_16s(IppsDCTLifterState_16s** pDCTLifter, int lenDCT, const Ipp32f* pLifter, int lenCeps)

there are two somewhat similar variables 'lenDCT' and 'lenCeps', the latter stamds for the amount of features being extracted, I guess that there might be cases where lenDCT is larger then lenCeps (otherwise there would have been only a single arg and not two ), on the aurora samplenumDCT is set to 23, why does more DCT coefficients then desierered features are needed? I guess the answer is mathematical in nature.

Any help would be appreciated.

Link Copied

0 Replies

Community support is provided during standard business hours (Monday to Friday 7AM - 5PM PST). Other contact methods are available here.

Intel does not verify all solutions, including but not limited to any file transfers that may appear in this community. Accordingly, Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade.

For more complete information about compiler optimizations, see our Optimization Notice.