W3C home > Mailing lists > Public > public-xg-audio@w3.org > July 2010

Re: Picking out the channels of an audio source

From: Chris Marrin <cmarrin@apple.com>
Date: Wed, 14 Jul 2010 13:46:21 -0700
Cc: public-xg-audio@w3.org
Message-Id: <8ECB59AF-9231-4337-820D-6CFDC7EB1841@apple.com>
To: Chris Rogers <crogers@google.com>

On Jul 14, 2010, at 12:20 PM, Chris Rogers wrote:

> Hi Chris,
> 
> There isn't anything currently, but we could add a ChannelSplitterNode and ChannelMergerNode.
> 
> The ChannelSplitterNode would have a single input, and multiple outputs (one per channel).  For example, in the stereo case it would take a stereo input, with two mono outputs which could be routed in strange ways (ping-pong delay effects for example).
> 
> The ChannelMergerNode would have multiple inputs and combine all the channels into a single output.  In the simplest case, it would accept two mono inputs with a single stereo output.
> 
> We would have to be careful with the ChannelSplitterNode, because its number of outputs could change dynamically simply by connecting a different source (with different number of channels) to its input.

Yeah, all this sounds fine. It should extend to multi-channel (e.g. AC3) sources. It would be an interesting experiment to do some simple 3D spatialization using this technique. You could also add a subwoofer channel for explosion effects in games. Is it reasonable to do AC3 encoding from 6 separate channels in real time (not in JS, but in native code)?

-----
~Chris
cmarrin@apple.com
Received on Wednesday, 14 July 2010 20:46:53 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 20 September 2010 18:33:55 GMT