CEFT: Difference between revisions

From XiphWiki
Jump to navigation Jump to search
(Some ideas about CEFT)
 
No edit summary
Line 4: Line 4:


=== Overlapped FFT not critically sampled ===
=== Overlapped FFT not critically sampled ===
In the current implementation, we encode 4/3 times more samples than necessary because we use 256-point FFTs with 64 samples overlap.


Ideas:
Ideas:
Line 10: Line 12:


=== Non-harmonic signals (i.e. music) ===
=== Non-harmonic signals (i.e. music) ===
CEFT only works on speech because most of its coding efficiency is provided by the pitch predictor.


Ideas:
Ideas:
Line 17: Line 21:


=== Sparse spectrum ===
=== Sparse spectrum ===
CEFT tends to have musical noise, especially at high frequency when there are very few bits/bin.


Ideas:
Ideas:
* Use a "rotation matrix"
* Use a "rotation matrix"
* Prediction from lower frequencies
* Prediction from lower frequencies

Revision as of 20:17, 21 November 2007

CEFT

Problems with current implementation and possible solutions

Overlapped FFT not critically sampled

In the current implementation, we encode 4/3 times more samples than necessary because we use 256-point FFTs with 64 samples overlap.

Ideas:

  • Use an MDCT instead of the FFT
  • Do extrapolation on the input and use a wider FFT. Then optimise the search only for the "real" samples

Non-harmonic signals (i.e. music)

CEFT only works on speech because most of its coding efficiency is provided by the pitch predictor.

Ideas:

  • Sinusoidal prediction
  • Use two (or more) pitch periods and choose one for each bin/band/whatever
  • Use two (or more) pitch periods at the same time and use energy conservation to keep everything stable.

Sparse spectrum

CEFT tends to have musical noise, especially at high frequency when there are very few bits/bin.

Ideas:

  • Use a "rotation matrix"
  • Prediction from lower frequencies