Disklavier

I am trying to understand this concept of digitally produced tones 
generating harmonics within the concepts of what I know, so please bear with 
me for taking this on again!

I know that when "audio" tuning devices first came on the market, they were 
of no use to piano tuners because they did not generate "beats".  Sine waves 
do not generate beats (like what you see on an oscilloscope).  Beats are 
wha-whas that you hear when 2 notes are played on your piano.  Aural piano 
tuners listen to these beats and define partials in beats per second (bps).  
For instance, if you play an octave or a twelfth, you should get no beats or 
1/2 beat per second.  A fourth (is contracted slighly) should have a 
"pregnant" beat, and a fifth (expanded)can have a very slow beat.  Piano 
tuners are always arguing about how "pure" these fifths should be!  The 
easiest intervals to hear beats are the major thirds and the tenths.  The 
major third F3 to A3 (the F and A below middle C) should beat about 7 beats 
per second.  That ratio is easier to hear if you also play F2.  The interval 
between F2 and A3 is a 12th.  Anyway, 12ths in the tenor section of the 
piano are a good way to hear beats.

Now the problem was, that if a device gave you a wave, say A4 at 440 Hz, 
which is the standard, international pitch (almost the dialtone on my 
telephone)and you played F3, you would not hear any beats because sine waves 
do not generate any beats. So, (again, my abysmal LACK of physics 
knowledge)they "clamped" these sine waves and made square waves.  The 
clamping was the average high or low point of the wave before it curved.  So 
now, instead of getting those graceful curves on sine waves, you got 
rectangles, but square waves generate beats.

So, are digital audio waves sine waves or square waves?  Perhaps I can 
understand why a song played on my piano might generate harmonics, but would 
the same song played on an electronic piano do the same thing?  Would the 
tone generator waves be different from the piano string waves?

Carol Beigel

>From: "Terry Holek" <tjholek@...>
>Reply-To: disklavier@yahoogroups.com
>To: <disklavier@yahoogroups.com>
>Subject: Re: [disklavier] WAV to MIDI
>Date: Tue, 24 Apr 2001 23:03:39 -0400
>
>Carol,
>Your assumption that digital tone modules do not produce harmonics in the
>tones they generate is incorrect (I believe this is what you refer to as a
>"quiet" WAV file).  Most common (realistic sounding, non FM-based) digital
>tone modules are sample based.  They take raw samples, apply some filtering
>and manipulation of the data, and re-produce the desired sounds.  The point
>is that the raw sample will definately include many of the harmonics you
>talk about, and thus the final output.  Even the more advanced acoustic
>modelling techniques (non-sample based) will re-create as much of the
>harmonic content as possible.
>
>If one could do an FFT analysis of a portion of a WAV file, and just map
>notes to the frequencies observed, things would be simple. But, as has been
>discussed, the problem is much more complex.
>
>Thanks for the info on tuning.
>
>Terry
>
>----- Original Message -----
>From: "Carol Beigel" <carolrpt@...>
>To: <disklavier@yahoogroups.com>
>Sent: Saturday, April 21, 2001 8:42 PM
>Subject: Re: [disklavier] WAV to MIDI
>
>
> > I agree with you that these programs are still in their infancy - in 
>fact
> > I'm surprised they have come this far.  I also liked Akoff's Composer 
>the
> > best and will probably throw the $29 at it to at least get one-voice 
>MIDI
> > from wave.
> >
> > I know that harmonics are a problem in this conversion.  What surprises 
>me
> > though (and here I will show my abysmal understanding of physics)is that
> > harmonics are a problem in digitally generated tones.
> >
> > I am a piano tuner, so I understand all about the harmonics generated 
>when
> > when a string is excited.  In fact, we piano tuners use these harmonics 
>to
> > tune.  We listen to the coincidental partials, i.e. the bass section is
> > usually tuned as 6:3 octaves, meaning that the 6th partial of the bottom
> > note should be beatless with the 3rd partial of the top note. For the
>octave
> > C3 (the C below middle C (which is C4) to C2, one would listen to the G
> > above middle C (which is G4).
> >
> >   Notes on the piano are named for the octave in which they appear i.e.
>the
> > first C on the keyboard starting from the left is C1, middle C is C4 and
>the
> > last note, #88 is C8, therefore the G above middle C is G4, and the 3
>notes
> > below C1 are Bzero, B-flat zero and the first note on the piano is A0.
> >
> > The harmonic progression of partials is:
> >
> > 1.  Fundamental (the note you play)
> > 2.  Octave  (the octave above the note you play)
> > 3.  Fifth  (the fifth above the octave)
> > 4.  Fourth
> > 5.  Major 3rd
> > 6.  Minor 3rd  (the note two octaves above the one you play)
> > 7.  Seventh
> >
> > Basically, what all this means is that the lower notes, the bass 
>section,
> > generate a lot more partials that people can hear than the higher notes.
> > The nice thing about owning a Yamaha piano is that they are about the
> > cleanest harmonics you can buy!
> >
> > I think I will try turning down the volume and just stick with trying to
> > convert piano audio files to MIDI for the moment!  If anyone can explain
>to
> > me why a quiet wave file will generate harmonics, let me know.
> >
> > Carol Beigel
> >
> >
> >
> > >From: mick hamer <mick@...>
> > >Reply-To: disklavier@yahoogroups.com
> > >To: disklavier@yahoogroups.com
> > >Subject: Re: [disklavier] WAV to MIDI
> > >Date: Sat, 21 Apr 2001 17:30:46 +0100
> > >
> > >There are quite a few wave to midi programs on the web that are 
>designed
> > >to tackle polyphonic music--such as a piece recorded on the piano. 
>There
> > >some more web address at the bottom of my email. Sadly none of them
> > >works that well (and I have tried at least demo versions of all of
> > >them). For my money, the best of the bunch is Akoff's Composer 2.0. But
> > >I wouldn't buy one without trying a demo version first.
> > >
> > >Converting wave to midi is a very difficult computational problem. If
> > >you play middle C on the piano, you not only get the frequency of 
>middle
> > >C, but also several harmonics--a softer C an octave above, a still
> > >softer G above that, C two octaves above and a few more minor
> > >vibrations. Play a simple three-note chord of C and there will be well
> > >over a dozen different frequencies. The main problem for the computer 
>is
> > >sorting out softly played notes from these harmonics. The programs have
> > >different ways to doing this, but basically if you turn up the
> > >sensitivity to capture more right notes you also get a lot of wrong
> > >ones. Turn the sensitivity down and you lose a lot of right notes. 
>There
> > >are also other problems, such as the difficulty all the programs have 
>in
> > >detecting bass notes. If you want to get a feel for some of the 
>problems
> > >the Audiotomidi program is one of the better ones, and it is free.
> > >
> > >The Official alt.music.midi FAQ has list of programs and a rather
> > >pessimistic view of the area. A couple of years ago the only wave to
> > >midi programs around were monophonic--they could only tackle single
> > >notes. Personally I think the programmers have made tremendous progress
> > >since then.
> > >
> > >Anyway that's enough waffle: here's a few websites. Good luck.
> > >http://www.intelliscore.net/
> > >http://www.pluto.dti.ne.jp/%7Earaki/amazingmidi/
> > >http://www.btinternet.com/~irshatwell/WaveGoodbye/software.html
> > >http://andreenk.chat.ru/english/widi.htm
> > >http://www.asahi-net.or.jp/~ri7h-obt/htdocs/soft/e_gama.html
> > >http://www.midi.ru/audiotomidi/ (This is freeware)
> > >And the Official alt.midi.music. faq are at 
>http://home.sc.rr.com/cosmog
> > >ony/ammfaq.html
> > >
> > >Mick Hamer
> > >
> > >  Carol Beigel <carolrpt@...> writes
> > > >Someone inquired a while back about converting MIDI generated files
>from
> > > >their Disklavier to wave files so they could put them on CD.  I was
> > >cruising
> > > >the www.sharewaremusicmachine.com site (click on Notation) and found 
>3
> > > >interesting programs that might interest this list.
> > > >
> > > >TiMidity++ v2.10.2 is a MIDI to WAVE converter and player that uses
> > >Gravis
> > > >Ultrasound-compatible patch files to generate digital audio data from
> > > >General MIDI files.  This program is freeware.
> > > >
> > > >AmazingMIDI 1.60 transcribes WAV to MIDI. (REad the text file to make
> > >sure
> > > >your .wav are sampled correctly)  Let's you save a 30-second MIDI 
>file
> > >using
> > > >the demo, but the full version only costs $29.
> > > >
> > > >Akoff Music Composer 2.0 - is music recognition software which 
>performs
> > >WAV
> > > >to MIDI conversion on polyphonic music.  Full version costs $29.
> > > >
> > > >None of this software works perfectly without editing, so learning 
>how
>to
> > > >edit scores would be most useful.  I understand the full version of
> > >Cakewalk
> > > >ScoreWriter on costs $39 these days, but make sure the soundcard on
>your
> > > >computer is at least 32-bit. (16-bit soundcards will crash your
>computer
> > > >using ScoreWriter!)
> > > >
> > > >Carol Beigel
> > > >
> > > >
> > > >
> > > >_________________________________________________________________
> > > >Get your FREE download of MSN Explorer at http://explorer.msn.com
> > > >
> > > >
> > > >To Post a message to the group, send it to:   disklavier@...
> > > >
> > > >To Post a private message to Todd Muncy, the group's founder and
> > >moderator, send
> > > >it to:
> > > >disklavier-owner@egroups.com
> > > >
> > > >To reach our group's web site go to:
> > > >http://egroups.com/group/disklavier
> > > >
> > > >To visit Todd's family web site that contains some fun disklavier
>content
> > >among
> > > >other things, go to:
> > > >http://MuncyFamily.com
> > > >
> > > >THINKING OF LEAVING THE GROUP?
> > > >If you are thinking of unsubcribing because you are getting too much
> > >mail, go
> > > >the the web site and change your email delivery option instead.  That
> > >will fix
> > > >the problem, while maintaining your access to the group.  If you 
>insist
> > >on
> > > >leaving us completely send a blank email to:
> > > >disklavier-unsubscribe@...
> > > >
> > > >Know someone who wants to join?  Have them send a blank email to:
> > > >disklavier-subscribe@...
> > > >
> > > >
> > > >Your use of Yahoo! Groups is subject to
>http://docs.yahoo.com/info/terms/
> > > >
> > > >
> > >
> > >--
> > >mick hamer
> >
> > _________________________________________________________________
> > Get your FREE download of MSN Explorer at http://explorer.msn.com
> >
> >
> > To Post a message to the group, send it to:   disklavier@...
> >
> > To Post a private message to Todd Muncy, the group's founder and
>moderator, send it to:
> > disklavier-owner@egroups.com
> >
> > To reach our group's web site go to:
> > http://egroups.com/group/disklavier
> >
> > To visit Todd's family web site that contains some fun disklavier 
>content
>among other things, go to:
> > http://MuncyFamily.com
> >
> > THINKING OF LEAVING THE GROUP?
> > If you are thinking of unsubcribing because you are getting too much 
>mail,
>go the the web site and change your email delivery option instead.  That
>will fix the problem, while maintaining your access to the group.  If you
>insist on leaving us completely send a blank email to:
> > disklavier-unsubscribe@...
> >
> > Know someone who wants to join?  Have them send a blank email to:
> > disklavier-subscribe@...
> >
> >
> > Your use of Yahoo! Groups is subject to 
>http://docs.yahoo.com/info/terms/
> >
> >
>

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com