W3C home > Mailing lists > Public > public-audio@w3.org > October to December 2012

Re: about the AnalyserNode

From: Chris Rogers <crogers@google.com>
Date: Fri, 14 Dec 2012 10:18:17 -0800
Message-ID: <CA+EzO0mPB70visB9EJGp6x9UZr6hx3zkr8AuGwK7fuRRVVLkeg@mail.gmail.com>
To: Jieun Oh <jieun5@ccrma.stanford.edu>
Cc: public-audio@w3.org
On Tue, Dec 11, 2012 at 3:19 PM, Jieun Oh <jieun5@ccrma.stanford.edu> wrote:

> Hello,
>
> I am relatively new to exploring the Web Audio API, and I just joined this
> list... so I apologize in advance if my questions have already been covered.
>
> I was curious exactly what is done by the AnalyserNode in order to produce
> the frequency data. For instance, what type of window is used? Is it
> possible to apply zero-padding? Is it possible to specify hop-sizes?
> Finally, how does the smoothingTimeConstant affect the output?
>

Jieun, a Blackman window is used at exactly the size of the FFT (no zero
padding).  This node is only intended for very common basic visualization
and is not meant to be the front-end for a phase vocoder analysis (with hop
sizes, etc.).  For more advanced analysis, you can use a
ScriptProcessorNode and control every aspect yourself.

The smoothingTimeConstant is a form of first-order low-pass filtering,
where some portion of the previous analysis frame is blended with the
current frame.


>
> Thank you in advance for your response (and thanks for this great API!).
>
> Best, Jieun
>
Received on Friday, 14 December 2012 18:18:44 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:50:04 UTC