Replace ScriptProcessorNode with AudioWorkletNode in p5.SoundFile and p5.Amplitude #373
Conversation
oshoham
commented
Jul 10, 2019
oshoham
commented
- Replace ScriptProcessorNode functionality with equivalent AudioWorklet processors in p5.SoundFile and p5.Amplitude
- Ensure that a global p5 instance is created and initialized before the test suite runs, so that all AudioWorklet modules are loaded before the tests run.
| this.processor = this.audiocontext.createScriptProcessor(this.bufferSize, 2, 1); | ||
| this._workletNode = new AudioWorkletNode(this.audiocontext, processorNames.amplitudeProcessor, { | ||
| outputChannelCount: [1], | ||
| parameterData: { smoothing: smoothing || 0 }, |
There was a problem hiding this comment.
as a potential improvement, we could check the range here, as in the smooth method. But we could also just let the parameterDescriptors handle enforcing (and choosing whether to notify the user) about the range.
| defaultValue: 0, | ||
| minValue: 0, | ||
| maxValue: 1, | ||
| automationRate: 'k-rate' |
There was a problem hiding this comment.
inspecting this in Chrome 75.0.3770.100, somehow it has become 'a-rate'. Confusing, but looks like you're following the spec https://webaudio.github.io/web-audio-api/#ref-for-enumdef-automationrate%E2%91%A1
There was a problem hiding this comment.
@oshoham seems like this is related to a bug in the polyfill:
In Firefox and Safari (using the ScriptProcessorNode fallback) the smoothing param is being passed to the process method as a 4096 frame Float32 buffer, rather than as a number. This is causing NaN values when we attempt to smooth.
I think the best way around this for now is to make Smoothing one of the processorOptions that updates on a message, like toggleNormalize does. But what do you think?
There was a problem hiding this comment.
yeah, I see the bug, it looks like the polyfill is assuming that all worklet parameters are a-rate: https://github.com/GoogleChromeLabs/audioworklet-polyfill/blob/d6f1a14dacc597876b1a5f9cd8719729a131b585/src/index.js#L96
I think your suggested approach makes sense - another possible solution is to treat the smoothing parameter as a-rate (i.e. a Float32 buffer of values) and update the processor code accordingly. This seems more in line with the intended use case of AudioParams, although probably less efficient for non-AudioWorklet browsers. What's your opinion?
| defaultValue: 0, | ||
| minValue: 0, | ||
| maxValue: 1, | ||
| automationRate: 'k-rate' |
There was a problem hiding this comment.
@oshoham seems like this is related to a bug in the polyfill:
In Firefox and Safari (using the ScriptProcessorNode fallback) the smoothing param is being passed to the process method as a 4096 frame Float32 buffer, rather than as a number. This is causing NaN values when we attempt to smooth.
I think the best way around this for now is to make Smoothing one of the processorOptions that updates on a message, like toggleNormalize does. But what do you think?
### New - AudioWorklet replaces the deprecated ScriptProcessorNode (when available) in p5.SoundRecorder, p5.Amplitude, and p5.SoundFile, as part of @oshoham's GSoC project https://github.com/processing/p5.js/blob/master/developer_docs/project_wrapups/orenshoham_gsoc_2019.md #369 #373 - p5 library compiled with webpack so it looks a little different and we had to remove comments that were tripping up YUIDoc here in the p5.js repo ### Fixes - p5.SoundFile: stop() won't stop a soundfile loop https://github.com/processing/p5.js-sound/issues/401 - p5.SoundFile: addCue() not triggering function calls https://github.com/processing/p5.js-sound/issues/371 - p5.SoundFile: jump() only called if soundfile is playing #404 - p5.SoundFile: save() bugfix #406 - remove bad console call (#378)
- improved inline documentation - all examples start audio on user gesture processing/p5.js-sound#403 - AudioWorklet replaces the deprecated ScriptProcessorNode (when available) in p5.SoundRecorder, p5.Amplitude, and p5.SoundFile, as part of @oshoham's GSoC project https://github.com/processing/p5.js/blob/master/developer_docs/project_wrapups/orenshoham_gsoc_2019.md processing/p5.js-sound#369 processing/p5.js-sound#373 - p5 library compiled with webpack so it looks a little different and we had to remove comments that were tripping up YUIDoc here in the p5.js repo - p5.SoundFile: stop() won't stop a soundfile loop https://github.com/processing/p5.js-sound/issues/401 - p5.SoundFile: addCue() not triggering function calls https://github.com/processing/p5.js-sound/issues/371 - p5.SoundFile: jump() only called if soundfile is playing processing/p5.js-sound#404 - p5.SoundFile: save() bugfix processing/p5.js-sound#406 - remove bad console call (processing/p5.js-sound#378)
