- Aug 07, 2010
-
-
Jean-Marc Valin authored
Decision not yet implemented in the encoder
-
-
- Aug 06, 2010
-
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
after inter-band prediction
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
This saves almost 200 bytes of code size.
-
dump_modes() was re-using a loop index for a loop nested inside the main one, so it would only output the struct for the first one.
-
- Aug 05, 2010
-
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
- Aug 04, 2010
-
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
- Aug 03, 2010
-
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
- Aug 02, 2010
-
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
- Jul 29, 2010
-
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
- Jul 28, 2010
-
-
Jean-Marc Valin authored
Removed useless code in transient_analysis()
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
- Jul 27, 2010
-
-
Use a closed-form formula for the search instead. This requires an integer sqrt, so it is not actually closed-form, but the number of iterations is O(qb) instead of O(2**qb).
-
Use a closed-form formula for the codebook parameters instead.
-
Jean-Marc Valin authored
Also convert the stereo split code to use log(N)/2 as the bit allocation offset
-
The old code allocated too many fine bits to large bands. New allocations were derived from by numerical optimization using quantization MSE sampled from Laplacian distributed random data to within +/- 1 bit for N=2...160 and bits per band from 0 to 64. Those allocations could be modeled with only minor errors using a simple offset of 19/8+log2(N), with no bits spent on fine energy when there would not be enough bits remaining to code a single pulse. However, PEAQ testing suggested an offset of 14/8 was better, and that it was always worth spending at least one bit on fine energy.
-
- Jul 25, 2010
-
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
- Jul 24, 2010
-
-
Jean-Marc Valin authored
-