audio table, embed multiple samples?

I've created a whole bunch of mathematically generated single cycle waveforms and I want to embed them in an audio table so I can cycle through them, but just not sure how to do it.
Any help would be much appreciated, thanks.

 

There is as usual a few different ways to approach this. I'm not aware of a straightforward procedure for imorting multiple files to an audio table so a couple of suggestions from the top of my head would be:

1. Single row offset (easy import, a bit more hassle with playback). If you know the exact length of your waveforms in samples, you can put them after eachother in a .wav and import them that way. Then you will have to determine which one to play by setting the start and end of the waveform in the playback clock/counter.

2. One waveform per row (more hassle with importing, but easier to adjust playback clock/counter). This way, you use two audio tables. One source, one target. Load your sample in the source, then record it to the target. For the next sample, add one to the WY port. Repeat for each waveform. Then you can select the waveform by changing rows (you will still have to set playback length by the playback clock/counter, but all should then start at 0 (or was it 1?)).

 

Thanks magneson

 

There is already a device built for this is the UL,

http://co.native-instruments.com/index.php?id=userlibrary&type=0&ulbr=1&plview=detail&patchid=6915

it will allow you to assemble wavetables in audio tables.

also, you can try this thing I built...

http://co.native-instruments.com/index.php?id=userlibrary&type=0&ulbr=1&plview=detail&patchid=13586

its got a 16 step wavetable setup with 5000 available waves, it might already cover the waves you want.

Good luck,

Michael

 

Thanks Michael, just what I was looking for, I've already created my waveforms just wasn't sure how to embed them in a table, the 'wav osc' ensamble you linked is just what I was looking for. Thank you for saving me a world of pain.
Your synth looks like a real sound designers dream.

 

I have the same problem .
I creat a long audiofile of 5120 samples long in an external editor ( consist of 10 * 512 samples )
When loading into an audiotable ...I want to set the legth to 512 ..assuming I can sweep trough ...no dice , the portion is trimmed of
Setting length to 5120 gives me the long table ...thus al 10 waveforms ..also not what I want
So I don't know of a convenient way to create a long audiofile in an external editor , importing it in the audiotable and making divisions ( division legth = sample lenght of each sample ).

 

here's a basic method for wavetables, it'll alias, but most will.

http://www.nireaktor.com/reaktor-tutorials/morphing-wavetables-in-reaktor/

if you have a single sample with multiple wavetables you simply change the amplitude of the ramp oscillator, and add the ramp oscillator output to a value the is the start of the wavetable, in samples.

 

Is it possible to keep wavetables from aliasing?

 

i believe so, if you use a windowed sinc filter, and the original content is bandlimited in the first place.

i never bothered to check it out but the process is outlined in the very useful digital sound generation:

http://courses.cs.washington.edu/courses/cse490s/11au/Readings/Digital_Sound_Generation_1.pdf

starting on page 9, section 1.4

i believe i have an article on creating generic sinc filters up on the nireaktor.com site, if you search for the series on oversampling. applying it properly to a stored set of samples would take some work and is something i haven't tried though.

if you wanted it to be an audio table, i guess the 'easiest' way would be to set the polyphony to the number of taps in the filter and have each voice process one data point.

 

Table oscilator : http://www.native-instruments.com/fr/community/reaktor-user-library/entry/show/7441/
Table filter : http://www.native-instruments.com/fr/community/reaktor-user-library/entry/show/7443/

You'll get everything mentioned in this thread.

Best regards.

 

Since these UL links are not working any more is there a way in general to find what's referred to with all such out-dated links? And in this case, Michael, can you update or provide the names for searching? Many thanks!

 

the second link was to Hyroglyphicus.

http://www.native-instruments.com/en/reaktor-community/reaktor-user-library/entry/show/7513/

the first link to the wavetable builder tool I cannot find anymore, I will look again later.

 

Excellent, thanks; loving Hyro., btw downloaded everything from your search page.

 

I found the other link here...

http://www.native-instruments.com/en/reaktor-community/reaktor-user-library/entry/show/4981/

its called waveosc, it lets you assemble and build wavetables.