In short, I want to record presentation audio, create time markers for that audio in a database, and then provide marker-navigation of that audio content from a web page. What technology (e.g. HTML5 Audio, RTMP) can support this?
My requirements in more detail:
- quickly navigate to server-side marked points in server-stored audio, from the browser.
- avoid any proprietary client-side software, such as Silverlight; although due to penetration Flash is acceptable, and future-looking standards like HTML5 media are acceptable, provided it ships with the latest browsers.
- prefer to leave the 30-50 minute audio files un-split rather than pre-splitting on the selected markers; so that the markers can be seamlessly changed later.
- like to keep licensing cost minimal; although single-purchase licensed server-side technologies are fine.
- prefer to do most of this from IIS, where I have most experience. however, a parallel streaming server such as Adobe with Windows APIs is acceptable.
Here are my best guesses on a solution so-far:
- the presentations will be compressed and stored in mp3 files (but really, any advice on an easily seekable format for speech recording is welcome).
- the client will play a unicast stream rather than download file chunks (although TBR, below, challenges this assumption)
- HTML5 is not ready, so flash will be required at the client
- IIS Media Services is no-go as it requires Silverlight for seekable audio
- The leading products in this space, such as Adobe Media Server 5, are probably large kits with their focus on video media. I can probably find a more focused tool like Icecast to reliably do what I need.