[Feature Request] Extend the `createPeriodicWave` API to accept time-domain arguments. about web-audio-api-v2 HOT 9 CLOSED

This would also be a good excuse for providing access to an FFT routine, couching it in terms of creating the necessary arrays to be used for a PeriodicWave. Then, combined with an OscillatorNode and PeriodicWave, you would have access to FFT and IFFT routines, in a roundabout way. :-)

from web-audio-api-v2.

Looping an AudioBufferSourceNode might work, but it will behave differently from a PeriodicWave created from the time-domain data. In particular, setting the frequency and detune will probably produce a nicer waveform, and it will be band-limited. Don't think AudioBufferSourceNode detune will give band-limited signals.

from web-audio-api-v2.

You're looking for AudioBufferSourceNode and its loop, playbakcRate and detune attribute I believe.

In any case, I'm transferring this to the v2 repo, where we're handling feature requests.

from web-audio-api-v2.

Thanks, @padenot. I'll look into using AudioBufferSourceNode. Thanks for transferring this. I didn't know about the v2 repo.

from web-audio-api-v2.

Sorry. I thought I was closing the issue for the other repo.

from web-audio-api-v2.

Don't think AudioBufferSourceNode detune will give band-limited signals.

It does in Gecko (at the expense of being slightly slower to compute).

from web-audio-api-v2.

Teleconf: This can all be done without adding a new feature. It just requires an FFT for which there are lots of libraries out there that would work.

Maybe we can add a small document/wiki page explaining how it can be done.

from web-audio-api-v2.

Virtual F2F:

• Like other issues, this needs up to date benchmark on FFT speed in WASM, because it's well possible that this is perfectly doable in script today
• Older benchmarks exist, but this was before WASM and before WASM with SIMD, so it's time to do a new benchmark
• If performances are acceptable, this can be closed and AudioBufferSourceNode can be used instead.

from web-audio-api-v2.

AudioWG virtual F2F:

• This is doable today, closing.

from web-audio-api-v2.