Moved README.TI-DSP to main dir
[speexdsp.git] / TODO
1 Allocator override (speex_lib_ctl)
2 Better error handling
3 Fixed-point:
4   - Wideband
5   - Initialization
6   - Jitter buffer
7 Echo cancellation:
8   - Int array for denoiser hooks
9 Denoiser:
10   - Smooth gain (remove musical noise)
11   - Better noise adaptation
12   - Do some tuning
13 AGC:
14   - Use median filtering instead of "non-linear mean"
15
16
17
18 Features
19 -Add maximum/minimum bit-rate control for VBR
20 -Improve error handling (with perror-like call?)
21
22 Long-term quality improvements
23 -Improve perceptual enhancement (including wideband)
24
25 Standards
26 -Complete Speex RTP profile
27 -MIME type registration
28
29 ideas:
30 Peelable stream (double codebook, higher bands, stereo)
31 LPC from spectral domain
32 Better psycho-acoustic model. Masking curve from Vorbis?