I'm working with the web audio API, I succeed listening to microphone and get data with the analyser (getFloatFrequencyData
and getFloatTimeDomainData
) but how can I get the original sample data before being modify/processed by the analyser?
The sample datas are not modified by the analyser, what you get from the analyser are the time domain and frequency domain datas. If you want to access / modify your datas in real time, you may use a script processor node, see : http://webaudio.github.io/web-audio-api/#the-scriptprocessornode-interface.
If you want to record your data, you may use RecordJS : https://github.com/mattdiamond/Recorderjs.
Underneath, RecordJS uses a script processor node, so just look at the code for an example. Basically it's quite simple to use, you get input datas and set ouput datas.