1

In the Web Audio API you can use the convolver node to create a reverb effect. You can then mix a dry version of the signal with a fully wet version via gains nodes to control the amount of the effect. What I would like to know is if it is possible to modify things like reverb tail length and room size dynamically and the "best practice" approach to doing this. I assume this would involve the scriptProcessor node and the convolver node working together in some manner.

William
  • 4,422
  • 17
  • 55
  • 108

1 Answers1

3

A far easier way to do this would be to generate a new reverb impulse response (an exponential falloff of noise works as a cheap version), and crossfade between convolver nodes when you change the impulse response (you can't directly change the IR used for a single node without it glitching).

cwilso
  • 13,610
  • 1
  • 30
  • 35
  • 1
    Doing it this way just put an image in my head of a huge library of impulses all intended to work well with one another as a single library, and all stored in a gargantuan multi dimensional array with an accommodating algorithm to find the best one based on the combination of values the user plops in. I am curious if I am understanding this right ? – William Nov 18 '14 at 00:43
  • 2
    As @cwilso mentioned, you can generate artificial reverb responses as you need them, rather than having to store a huge number of them in advance. I recently started [an open source library](https://github.com/adelespinasse/reverbGen) to do this for you. It's still missing a lot of features that I'd like to add, but it works, and might save you some time. – aldel Nov 18 '14 at 15:21
  • I misunderstood his point. I didn't realize you could dynamically create the responses [the way you have in this](http://aldel.com/reverbgen/). I assumed when he said "generate new responses" he meant outside of the web audio api - and then to import them.Hence my remark about a giant multidimensional array to store them all :) – William Nov 18 '14 at 18:02