views:

74

answers:

2

I can't find anything through net searches, but is there any plans to make an API to generate audio chunks to be played in an HTML5 <audio> tag?

EDIT: this is the example::

PSEUDOCODE::
var music = new Song([Array of hertz levels or notes]);
var box = document.createElement('audio');
document.body.appendChild(box);
box.src = music.convert();
box.play();
+2  A: 

In theory, you can actually do this now -- set the source attribute of an audio element to a data URI which you build on the fly in your JavaScript. I've done this with embed elements and MIDI files, which works on machines with QuickTime or some other media helper installed.

The catch, of course, is that generating audio data is a bit more computationally intensive and it takes up more space (which might run afoul of the data URL size limit in certain contexts... IE and Mobile Safari come to mind). So you'd be limited to small clips, and you might be limited by start/stop times for the audio tag.

Mozilla has an audio buffer API they're working on, which could bring true programatically generated audio to the browser. In the meanwhile, though, it's either Data URI hacks, or plugin-type stuff like Flash (which got a programatic Audio Buffer in version 10) or Java.

One more note: I started a JavaScript Audio google group a few months back. There isn't a lot of discussion going on there yet, but I'd love it if enough people who were interested in music/audio/sound in the context of the browser signed up so that we can reach a critical mass and start having actual conversations on topics just like this.

Weston C
A: 

Mozilla have released an audio data API, which works in Firefox 4 betas and Minefield nightlies:

https://wiki.mozilla.org/Audio_Data_API

http://vocamus.net/dave/?p=1148

As I understand it, there is also a similar WebKit project.

A W3C Audio Incubator Group has been formed -- their mission is 'to explore the possibility of starting one or more specifications dealing with various aspects of advanced audio functionality, including reading and writing raw audio data, and synthesizing sound or speech'.

http://www.w3.org/2005/Incubator/audio/

Sam Dutton