🌱 audio worklets
In the last post, we've been filling fixed length audio buffers. While this is very simple to implement, it's a bit annoying when the audio is always stopping. To calculate samples infinitely, we can use an AudioWorklet. Now we've arrived at a typical bytebeat/floatbeat editor:
💡 hit ctrl+enter to update the code & ctrl+. to stop
((( // by stimmer (2011-10-03)
t*(4|t>>13&3)>>(~t>>11&1)&128|t*(t>>11&t>>13)*(~t>>9&3)&127
) & 255) / 127.5 - 1)/4
play | stop
The AudioWorklet is the lowest abstaction we get with the Web Audio API. It essentially gives us a function that fills a buffer of 128 samples over and over again, also running isolated from the rest of the page. At a sample rate of 44100/s, this means we fill a buffer every ~3ms (44100/128).
Here are some examples to try:
hello sine
slow sine (inaudible)
just intonation triad
hello saw
hello pulse
wandering sine
bytebeat 1
bytebeat 2
gladyouask - chirps
aks - TR-808-like hihat
show source (271 loc)
🌱 audio worklets
🌱 audio worklets
In the last post, we've been filling
fixed length audio buffers. While this is very simple to implement, it's a
bit annoying when the audio is always stopping. To calculate samples
infinitely, we can use an AudioWorklet. Now we've arrived at a typical
bytebeat/floatbeat editor:
💡 hit ctrl+enter to update the code & ctrl+. to stop
play | stop
The AudioWorklet is the lowest abstaction we get with the Web Audio API.
It essentially gives us a function that fills a buffer of 128 samples over
and over again, also running isolated from the rest of the page. At a
sample rate of 44100/s, this means we fill a buffer every ~3ms
(44100/128).
Here are some examples to try:
show page source
back to garten.salat
back to garten.salat