The short version of my question is:
How may I realize seamless audio capturing on a web-site and save it on the server?
"seamless" means not having to use an external program for recording and afterwards uploading the file.
Few constraints:
1) No Adobe or other commercially licensed tools
2) Runs at least on Chrome or Firefox of most recent version (that means also browser plug-ins come into play for me!)
3) no java
I am wondering why this feature isn't yet realized for HTML5 as it is specified as far as I know and is a very desirable one for sure!?
I am a bit confused regarding this quote:
Another area where Flash will remain is videoconferencing and recording audio or video. HTML5 has no method for controlling a user’s microphone or web cam. This means that Flash will continue to be used for conferencing applications and video chat programs.
http://www.jimcode.org/2011/07/html5-and-flash-why-flash-is-here-to-stay/
as it states that this feature isn't planned for HTML5!?
YUI Theater — Scott Schiller: “Probably, Maybe, No: The State Of HTML5 Audio” (51 min.)
Chrome 14 Beta Adds Native Client, Lion Support, Web Audio
Speech Input API Specification
There are two methods of handling this
The HTML5 speech input is currently supported by chrome (and maybe safari) as
<input type="text" x-webkit-speech />
In theory you can write a webkit-speech plugin for firefox and this would meet your chrome & firefox support criteria. In practice writing such a plugin is a beast of a job.
The WebRTC is supported by chromium and is experimental
Also it turns out webkit has navigator.webkitGetUserMedia (which is defined in the WHATWG specification) but I can't get it to do anything. You might need chromium for it to do something