Greetings;
I'm interested in sending voice quality data (bitrate=8) to HTML5 <audio>
elements using Liquidsoap. Unfortunately, it doesn't look like any of the mp3
varieties (fxp, vbr, abr) in Liquidsoap allow bitrate=8. I realize that speex
uses bitrate=8, but MP3 is preferred because IE, Chrome, FF all support it in
HTML5 (Windows 7).
MPEG2 III states that it supports this bitrate, but Liquidsoap must be MPEG1
III? Is there any way I can get this to work?
In our previous C++ version of this App I used Speex library and have that
audio source available in speex bytes. Am I correct in my assumption that the
only way to retrieve this speex data would be to use
add_decoder("speex-to-wav") and then
output.icecast(%mp3,bitrate=8,...,wav-input), or is there an easier method
(input.harbor)?
Oh, one more thing. In the C++ version, I split the data into speex headers and
data and sent them separately. I always sent the headers and used FFT to
calculate the Db level on the data portion, and only sent the 8K of data if it
was not silence. I tried to do the same thing in Liquidsoap with skip_blank().
Am I on the right track?
James A. Thompson
[email protected]<mailto:[email protected]>
------------------------------------------------------------------------------
_______________________________________________
Savonet-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/savonet-users
