What I need to do is, for example:

Take soundfiles, aka. tabplay~'s of varying sizes/lengths and shrink them all so they play over the same amount of time.

Example:

Soundfiles 1000ms, 500ms, 250ms long that all play completely in 125ms.

Increasing in pitch is fine, as well as, using externals is fine.

I figure between zexy, cyclone, and iemlib, one of them does it.
But I ran searches for "shrink", "compress", and "condense", and didn't find anything.

Thanks, in advance, for your help.

Peace.

-s

@svanya You can use [line~] or [line] or [vline~]. Set it to output 0-1 in 125ms. Use a multiplier appropriate for each table at its output (multiply the output by the table size).
For [line]...... as that is appropriate for the indexing ([tabplay~]).... you will need to set a fine grain.
I think it will be better to use [tabread~] and [vline~]
You could also use [phasor~] set to 8Hz and mute the audio as the output reaches 1........ but the problem is that it never does.
David.

@svanya said:

What I need to do is, for example:

Take soundfiles, aka. tabplay~'s of varying sizes/lengths and shrink them all so they play over the same amount of time.

I think cyclone/play~ is the easiest way to do this. Its "start" message is in the format:

• Time within the array where you want to start playing (ms);
• Time within the array where you want to stop playing (ms);
• Duration to play.

The playback rate is then automatically calculated from (end - start) / dur.

The problem with cyclone buffer~ / play~ is that (AFAICS) they don't handle the situation where the audio file being loaded is at a different sample rate from the soundcard. (Maybe they do, but if so, I don't see it.)

If you're sure that the sound files will match the DSP sample rate, then, no problem -- if one array contains 750 ms of audio and you want it to play in 250 ms, then:

[start 0 750 250(
|
[cyclone/play~ arrayname]

If you're not sure, then one solution would be some abstractions in my library (https://github.com/jamshark70/hjh-abs) that provide [value] variables with information about sound files loaded into arrays. [value xxxscale] is file's sample rate / system sample rate (btw DSP should be on before loading... I should fix that later).

(Edit: You may need to have "-lib cyclone" in the [declare] too.)

22-0810-sf-same-dur-play.pd

hjh

@svanya take a look at B14.sampler.rockerfella.pd in the /doc/3.audio.examples

Thank you, @whale-av @ddw_music and @balwyn, for the leads.

After looking into and testing cyclone/play~ it's going to work out great.

It's for a meta-metronome, percussion, sampler app I'm working on.

And am excited about the prospect that

1. I can count on all the samples for each of the tracks being a set length

and

2. the notion that by ramping up how fast it plays (say switching from 1/4 notes to 1/8ths) Will increase the pitch.

Conceptually Very appealing as it's based on the notion of taking a universal metro length and subdividing it 8 different ways once for each track.

Play~ in a way will add pitch to the whole originally "drum"-ish concept.

Thanks, again. I Very Much appreciate the (as always ) quick turn around on my question.

Peace through Music.
Love through Sharing
And...Vision through PD .

@ddw_music said:

(btw DSP should be on before loading... I should fix that later).

This should be fixed now in the github repository.

Sound file handling in Pd bothers me a bit -- harder than it needs to be. I can sort of see the MSP Theory and Technique of Computer Music hard line, "it's of benefit to learn the math" even for simple things like playing a sound file at normal speed, but where I teach, that frightens students away.

Anyway glad you found a solution!

hjh