Greetings,

Anyone done something like this?

[metro 10] [ADC~ 1] ; a microphone
| |
[ tabwrite~ adc1 ] ; microphone is sampled every 10 ms (~441 samples)

The table adc1 is then read by a tabread4~ which is looked up by a [phasor~]-[abs~]-[*~ 441] (forward and reverse, I'll explain)
The frequency is controlled by MIDI keyboard.

Awwww, now you see, it is a midi-controlled "synthesizer" using the sample of your voice for the waveform. Very simple program, please experiment if you haven't to discover the issues.

I have my code sloppily "double buffered" (two alternating tables) in a hopeless attempt to reduce distortion.

My hopes and queries:

1. eliminate distortion

2. set the start point of the table lookup to correspond to signal=0 and look up exactly one period of the waveform (crosses 0 again, resets at next 0.) This way the [abs~] would be unnecessary as it is a means of "smoothing the edges." It would also calibrate the pitch. The logic sounds simple enough but I was getting nowhere with the unsig~ "re"sig~ voodoo that I attempted. Any suggestions?

I am thinking that I would be better off sampling 64 samples at a time (1.514 ms or something.) Or more likely 256 or 512 for a full period. Forget about the waveform logic mentioned in number 2), the patch sounds badass as it is and that's all that matters.

Does anyone think that I could use tabosc~ or some exotic table object that may suit my purposes better? Being able to change tables would be nice, for starters.

Please try it out, it will only take 5 minutes to get it making noise, and it is an awesome patch if we get it working properly - much better than a vocoder - and if you don't believe me, try it. Just A E I O U to your heart's content.

I've been meaning to post for months, so a belated thank you to all whose work has frustrated me to the point I'm at now, and a thank you in advance for your help. I've been working on an all in one Techno patch that is very CPU cheap - I will get around to uploading it soon. If anyone out there has a MIDI controlled analog synth you will enjoy my ACID patch - if anyone out there likes break beats, you will enjoy my break looper, randomizer, pitcher, and keyboard controlled stustutter.

*If anyone isn't using a 96tic/bar master clock take that as my first contribution.

This most recent effort is an attempt at something more melodic, though I am dismayed by my 100ms latency on G4 1.33...Debian will be my savior I hope.

Happy projramming

great ideas there.

the attached patch follows most of what you suggested, with a little bit of pd knowledge.

the main thing i have exploited is the use of overlapping blocks. right click on a block~ object to get its help file and more info about that. basically, with a block overlap of 2 (default is 1), every process in the block is augmented by the same process triggered 50% of the way through the playback of the first block.

that takes care of a lot of stuff automatically - the crossfading between 2 tables, the phased-playback of the 2 tables, etc.

"2 tables?", you might ask, when you open the patch and see that there is only one table. Well, with the overlap, another table is automatically invisibly created to be read by the overlapped block.

http://www.pdpatchrepo.info/hurleur/vokoder2.pd

i did something very similar a long time ago, didn't get very far though and you I cant find it anymore. so I could well be lying, but anyway! That's all I have to say about it really.

Well, I do ask, mod, can you explain how the (block~) overlap works and how it is (automatically) implemented? I created this in a patch manually, and wondering if [block~] could have saved the day. - shall we start a new topic so future searchers can find the answer?

[--------][--------][--------] <--- block 1st overlap
----][--------][--------] <---- block 2nd overlap

i think it's like that with an overlap of 2. all SIGNAL operations occur twice, with the block start time offset on the 2nd overlap.

for overlap of 4, you have 4 threads, with only 25% block start offset between each successive overlap.

this is just my guess how it works though. i didn't read anything, and just figured it out by trial and error.

Well thanks for the suggestions. I will post my more complicated patch if/when I clean it up. I say "if", because I just found an object that makes the whole thing really easy - haven't manipulated it too much yet, but it seems promising.

[dynwav~]

So I've attached the simple result that basically does what I was trying to do probably cheaper and just...better. However, my patch at least maintains control over the pitch whereas the following patch you may notice that the pitch of your voice is reflected in the output...unless you mess with the block~ in the subpatch.

If you have a notebook it's as easy as headphones and an internal mic.

http://www.pdpatchrepo.info/hurleur/mic2.pd

Nevermind, that shit is terrible. Here is an update of the original patch that I never posted.

http://www.pdpatchrepo.info/hurleur/mic.pd