Middle – the tidalcycles python mashup

Uncategorized

So I redid the text to voice + poem junk generation + tidalcycles. I need to save supercollider to buffer so you can’t hear my keystrokes. BUT all the sound is made with the text to speech poem generator. Thats cool.

I feel like it starts to pick up at min 1 and you can hear me curse at like  min 3.

But what happened before….

I took a look to see why my text to voice clips in tidalcycles did not work as expected . After some googling I found this great documentation.  To Quote:

As you can hear, Tidal will keep triggering the sample each cycle, even if it’s very long. Even if you stop the pattern playing, you will still need to listen while the samples play out.

This is sort of what I thought I was hearing. So how do we fix that?

something like:

d1 $ sound “bev” # cut 1

repeats the same clip again and again. Also “hush” does not work as expected because it I think needs to wait for all the cycles to finish.

Something like

d1 $ slow BIGNUM $ sound “bev ~” # cut 1

Keeps the clip going throughout many cycles

and

d1 $ chop 32 $ sound "bev"
d1 $ striate 32 $ sound "bev"

Let you play a clip from the middle!

When I went back and listened to all the clips I realize the audio did not record! Riley at RC was correct.

Here are the original clips of the two poems:

I am sticking my config here so I remember:

Quarks.checkForUpdates(); Quarks.install("SuperDirt", "v1.0")

(
s.waitForBoot {
	~dirt = SuperDirt(2, s); // two output channels

        // load samples from multiple folders:
	~dirtloadSoundFiles("/Users/Parmenides/projects/RC/tsamples/
	~dirt.loadSoundFiles("/Users/Parmenides/projects/sounds/*"); 

	s.sync; // wait for supercollider to finish booting up
	~dirt.start(57120, [0, 0]); // start superdirt, listening on port 57120, create two busses each sending audio to channel 0
};
);
SuperDirt.start

Leave a Reply