Hello vorbis developer! I'm not very experienced in informatics, cause i'm only an audio engineer and sorry for my broken english but i have an idea... After the filterbank the maximum aplitude in each band decides about the nessesary bits, then the psychoacoustic model decides how much of the signal can be truncated without hearable noise. What do you think about quantising the remaining bits nonlinear with a logarithmic scale? This would share 25%. Nonlinear quantisation is used for instance in a-law or for the longplay-mode of DAT-recorders. The disadvantages of nonlinear quantisation will not apply here, because the 576 frequency bands are too narrow. The harmonics of the square waves are outside these bands and will be filtered out by the filterbank on the player side. The Result would be a better resolution on low amplitudes. The quantising noise on high levels would be masked thru the signal itself. I know, this is not MPEG-compatible in any way. Anyway what do you think? Would it work? ----------------------------------------------------------------- Sirko Zidlewitz sirko@track4.de Recording Studio track4 Tel:+49-3381-3159052 Fax:+49-3381-3159054 Magdeburger Str. 15, 14770 Brandenburg a.d. Havel, Germany --- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/
> After the filterbank the maximum aplitude in each band decides > about the nessesary bits, then the psychoacoustic model decides how > much of the signal can be truncated without hearable noise.Vorbis does not use critical bands or any subbanding.> What do you think about quantising the remaining bits nonlinear > with a logarithmic scale? This would share 25%.That's entriely up to the choice of codebook, and the codebook can be whatever the encoder decides. Ah, you mean quantizing time domain residual? There are hooks for time domain coding, yes, and Greg Maxwell has played with wavelet encoding this. However, this is not meant to catch inaudible residuals as much as it is to eliminate impulses and plosions that do not represent well in the frequency domain.> The Result would be a better resolution on low amplitudes.There are not really absolute amplitudes of any sort involved... Vorbis has an (obviously absurdly large) representation range of over 300dB. Encoding a low dB signal is not different than encoding a high dB signal. The psychoacoustics are designed to look at the range of the input signal, not how loud it would be from an arbitrary zero dB point. Monty --- >8 ---- List archives: http://www.xiph.org/archives/ Ogg project homepage: http://www.xiph.org/ogg/
Seemingly Similar Threads
- Line-In Encoding
- patents and separate entropy coding
- [PATCH] Moved the application of the pitch gain to (un)quant_bands(). This doesn't change anything to the codec, but will make the next changes easier.
- Status of s3tc patent in respect to open-source drivers and workarounds
- [PATCH] Pitch changes