Kenji's K5000 Message Board Digest - Overview K5000 Resources - Overview
The Eat at Joe's Kawai K5000 Message Board Digest
Resynthesizing Audio Files with Sounddiver


resynthsis through SoundDive
 Monday, 08-Dec-97 05:35:53

      Message: writes:

      I just got my k5000 and here's my first round o questions . . .

      I've been reading message posted to Analogue Heaven about the K5000 resynthesizing
      samples into additive patches. Is this true? How do you do it? I've got
      the Sound Diver K5000 edition that came with it and the Memory Exapansion and I
      can't figure it out.

      Also can you change the Resonance while holding down a note? The Resonance macro
      knob only seems to work after you retrigger a key, however if you
      assign Resonance as one of the destinations of the User Macro knobs then it seems
      you can modulate Resonance while holding down a note. Is this right?

      Is there a multi-mode on the thing. Is the only way to achieve multi-timbrality
      through setting up multi-patches?


      btw ->Music Central $995 with ME-1 and 4 patch disks K5000S 


Re: resynthsis through SoundDive
Monday, 08-Dec-97 17:22:15 writes:

     Hit the Import button on the ADD panel of SoundDiver for the resynthesis function.
     It works off wav files. Let me know if you get it to work--I've only tried it a
     few times, but I didn't get any useful results out of it.

     You're right about the resonance knob--I'll try that fix, thanks.

     Not sure if this is what you're asking, but: For Multi mode, remember to set the
     patch select mode on the midi screen of the system menu. (It isn't in front of me,
     but I think those are the right names.) In "norm" mode, a patch message will select
     a multi patch, but in the other mode ("sec"?) patch changes on up to four
     channels can change the individual patches within a multi patch. 


Re: resynthsis through SoundDive
Tuesday, 09-Dec-97 09:28:38 writes:

     Hello Michael,

     in the k5000 editor under additive synthesis
     you can import a .wav file. This file will be
     used for resynthesis. I have tried it out with
     vocals and it works ok.



sounddiver import from WAV experience?
Saturday, 17-Jan-98 19:06:38 writes:

     Hi everybody,

     I just started using soundiver to create starts
     for patches with the WAV import function:

     - The atack phase of some harminics looks very
     strange sometimes. I guess the best way is to
     use it is to select a stable part from a sample
     and add some dynamics later by hand.

     - Sometimes the sounds get very distorted. It 
     must be a very special kind of fourier trans-
     formation to turn a grand piano sample into
     a fart-like sound ;-)

     Anyone has had some experience with it ?

     Groet, succes, Bart.

     Bart van der Worp, 


Re: sounddiver import from WAV experience?
Tuesday, 03-Feb-98 20:34:11 writes:


     I haven't got a k5k myself yet, but I've read alot about them over the past
     couple of months. All I can say is that I don't know how the synth can produce
     realistic acoustic tones because you can't detune the actual harmonics within a
     source. This is crucial for acoustic emulation. I think a synth like the Z1 is taking
     a better approach for that sort of thing. That is why I've not gotten either one
     yet- I can't decide. 

     Anyway, it sounds like your computer is doing its best to translate the file, but
     the Kawai may not really be suited for that. Maybe that will require a system
     update, but how many updates can they really do? Imagine taking a tomato and
     throwing it through a door. On the Kawai, it's a screen door, and thus you
     end up only with tomato paste on the other side. Does this make sense? What do
     you think? 


Re: Re: sounddiver import from WAV experience?
Wednesday, 04-Feb-98 03:48:20 writes:

     Is it really crucial for acoustic emulation to be able to detune harmonics? I am
     just asking, I wonder if blindfold tests have been carried out that support this

     There are so many irregular aspects to acoustic sound that synthesizers may not
     ever be able to emulate all of them exactly. Therefore, one really should carry
     out blindfold tests in order to decide which ones are crucial.

     Having said this, I don't think that synthesizers *should* try to emulate acoustic
     sounds. What I am looking for, rather, is sound that has the finely structured
     fabric of acoustic sound, without necessarily emulating this or that instrument.
     I think the K5000 can do it - at least, I don't know what other synth currently on
     the market would be a serious competitor in these respects.

     I also considered a Korg Z1, but it seemed to me that the sound quality of the
     K5000 was as good or better. Given that the K5000 is also more flexible, I
     ended up buying this machine - even before taken the price into account. 



Re: Re: sounddiver import from WAV experience?
Wednesday, 04-Feb-98 16:00:40 writes:

     Also, although it uses a sort of workaround, you can detune individual harmonics.
     The K5000 allows for six different sources (a source is a fundamental and
     an overtone series) for each patch. Then, you can create a multi-patch which
     combines four different patches. This gives you a total of twenty four different

     What this means is that you could have the first source play the fundamental tone
     of the sound you are trying to create as well as the harmonic overtones.
     Then, you can detune other sources to get any other inharmonic overtone(s) you
     want. With twenty four different sources, each with a complete overtone
     series, you really could generate a very large number of inharmonic overtones
     if you wanted to (up to a maximum of 1536 sine waves! That makes me almost
     want to try it just for fun.)....

     Admittedly it would be much easier to do this if you could set the pitch of each
     harmonic or detune them, but detuned overtones definately *can* be done.

     Of course this won't help at all with the original message on this thread, because
     the wav converter doesn't do multiple source detuning. All I can say is try
     clipping out a "loopable" type clip of the piano note after the attack transients
     are all gone and the tone has entered a regular cycle. Then try to emulate the
     attack using other methods. I really think that's more what the Sound Diver wav
     converter is supposed to be for. Have you tried that yet?



Resynthesis limitations
 Monday, 03-Aug-98 14:00:07 writes:

      I'm very interested in understanding resynthesis,
      and would like to know your opinion.

      I think that, no matter how exact you can analyze
      a waveform, the problem will be how to code a
      changing spectrum (over time) into the K5K.

      There should be "no problem" in static waveforms,
      guess they can be reporduced exactly. But to
      model a changing spectrum you only have the
      chance to express this change in terms of
      attack/decay/loop of the envelopes of the
      harmonics or/and the formant filter.

      There is no way to store a whole bunch of wavesets which should become active at
      predeterminded times.

      Am I right or do I miss something?

      - Andy 


Re: Resynthesis limitations
 Tuesday, 04-Aug-98 04:37:56 writes:

      Take a look at SoundDiver's on-line help. (Open help from the ADD single editor
      and find the keyword 'resynthesis' in the help index). There is a short
      description of Emagic's method there.

      Jens Groh 


Re: Resynthesis limitations
 Sunday, 09-Aug-98 14:14:56 writes:

      The Morph function may allow you to at least move among 4 resynthesized parts,
      just off the top of my head.



Re: uses for the wav import function (sound diver)
 Thursday, 22-Oct-98 09:31:33 writes:

      When it comes to wav import I've found that the best wavs to use are 1 secound wav
      that have no frequency mod.

      For example, I use a program called WavGen to create an original 1 secound wav,
      by blending together squares, pulses, sin/cos, etc. Sometimes its a good idea to
      get a good balance of frequencies in this raw wav. You can always filter the ones
      you don't want out later. Once I'm happy with the wav I save it and import the
      buggar in. This has helded some great starting points for sound generation.

      As far as the differnt types of import I try out each one, and pick the one that
      best suits my taste.

      This aplication doesn't work well if you try and import an acoustic violin or
      guitar, etc. SO i stick to raw synth wavs.