On 22/12/2011 08:08, Richard Outerbridge wrote: > In article > <email@example.com>, > RichD<firstname.lastname@example.org> wrote: > >> On Nov 4, Martin Brown<|||newspam...@nezumi.demon.co.uk> wrote: >>>> He estimated music contains 40 bits/second entropy. >>>> How close is MP3 to that? >>> >>>> ___________________________________ >>>> I doubt surprised Shannon said that, and if he did its >>>> somewhere between meaningless and wrong. >>> >>>> CD quality mp3s are roughly equivalent to 178 kbps, >>>> over 4,000 times his estimate. But then you can >>>> encode a lot of sounds that most people would not >>>> consider music. And it stereo, so you can halve it >>>> if Shannon was talking about mono. >>> >>> Have you never seen sheet music? That is what >>> Shannon was estimating - the bitrate for describing >>> music in the abstract. >>> There are a finite number of notes, durations and >>> amplitudes in a classical composition. >> >> I suspect that's the case. >> >>> I suspect 40 bits/sec is still far too tight, >>> but a midi stream using a high end reconstruction >>> codec represents a pretty good >>> example of what is possible by way of compression >>> for *music* as opposed to voice or a random noise >>> stream. >>> >>>> To get a smaller figure for music, you have to >>>> define what subsets of sounds are music. Lots of luck. >> >> Not so hard, if you analyze samples of what we label 'music'. >> >>> I think that may have been his intention although >>> I don't recall seeing the 40 bit/s number >>> As I said if he did anything I think he was estimating >>> the information content of music in the already >>> concise form of an orchestral score. >> >> That appears to be the case. >> >>> I reckon at a bare minimum about 7 to the note, 8 to >>> amplitude, 6 duration, 5 to the instrument - and it >>> is already obvious that you cannot encode more than a >>> single note per second at this bitrate. >> >> You fail to account for the essential concept >> of correlation. >> >>> Can anyone provide a citation to this alleged paper >>> on music bitrate? >> >> >> I found the reference in John Pierce's book. >> It's not attributed to Shannon. My boner. >> He claims a humans can absorb information, >> e.g. reading or music, at 40 bits/second.
That sounds way too low. I can flash read this entire page in about second and there are a lot more than 40 bits of data in it. And the processing to do that much text recognition in real time would tax a super computer - that is why Captcha is used to test for humans!
>> He also cites a paper where it's claimed that >> sheet music contains 2.5 bits entropy per note.
Again sounds far too low even when you only allow for pitch, duration and amplitude (plus any changes in amplitude, incidentals, time signature or syncopation). Please can you post the citation?
I would only believe this claim iff they can demonstrate a real musical score compressed to the equivalent of 2.5bits/note. That is the acid test. Hand waving arguments and gut feel do not hack it.
>> This supports the MIDI model, as opposed to >> digitization of a recording. In that case, >> what is the bit rate? Though the question of >> chords arises, with much redundancy there. >> >> It also raises the possibility that recorded >> music might be adequately represented by a >> MIDI stream. > > IIRC, it's interesting to note that most modern media > compression codecs rely not so much upon what the raw > data can absolutely do without as they do upon what the > human ear or eye can be fooled by: they leave out what > you "cannot" hear or "cannot" see. Really, this is > nothing new: all screens of any sort are really just > optical illusions of one sort or another. The codecs > are just leveraging the limitations of the wiring of > the human brain.
Absolutely. There is no point in transmitting detail that the ear cannot hear or the eye cannot see. But you do have to be a little bit careful. The UK's new DAB audio bitrate is insufficient to support quality live broadcast of classical music and is distinctly inferior to either FM or off satellite SD or HD digital audio. The problems with DAB are sufficiently bad that it is hampering uptake of new DAB radios.