[Live-devel] Regarding MP3 Streaming

[Ross Finlayson]

>I was working on MP3 streaming code and what I understood is first 
>an MP3 frame is converted to ADU and then again this ADU is 
>converted to MP3 frame. Although I am still trying to understand the 
>complete flow but can you tell me why we are doing this conversion 
>from A -> B -> A . Is that just because to support trick mode or 
>something else.

(I presume you're referring to the "MP3AudioFileServerMediaSubsession" code.)

Yes, the conversion MP3->ADU->MP3 is done in order to support 'trick 
mode' - specifically, fast forward.  Because MP3 frames are not self 
contained, it is not possible to implement Nx speedup on a MP3 stream 
simply by extracting every Nth MP3 frame.  Instead, we need to first 
convert to ADUs, apply the scaling on the ADUs, and then convert them 
back to MP3 frames.

(For more information on MP3 ADUs, see 
<http://www.live555.com/rtp-mp3/> and 
<http://www.live555.com/rtp-mp3/rtp-mp3.txt>
-- 

Ross Finlayson
Live Networks, Inc.
http://www.live555.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.live555.com/pipermail/live-devel/attachments/20070217/303a6dc1/attachment.html