Need input on "seamless audio capturing on websites" - what is the state of the art?

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开发者_JS百科-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

The Capture API - W3C Working Draft 01 April 2010

HTML Media Capture - W3C Working Draft 14 April 2011

---

There are two methods of handling this

• HTML5 Speech inputs
• WHATWG Real time communications

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

---

The HTML 5 Device API is currently being developed, and it will allow native HTML/JavaScript access to input devices such as cameras and microphones.