> i thought that the chirp could evntually be an aliasing effect where > frequencis in the range above 11khz are mirroed into lower bands by the > sampling process. in order to rule that out i did some testing with encoding > a short wav file.. i implemented a downsample by 2 quick hack into the lame > source (takes the avearge value between 2 samples..) > > so you can write: > > lame -b 24 -s 44.1 --resample 22.05 -a waits.wav waits24.mp3 > > (but only if the resample frequency is exactly half of the input frequency.) > > i will try to pacht liveice so it understands that option and see what the > audio technican thinks about the quality now.. has any of you some > comparision of lame encoder vs. comercial encoders at low bitrates? One thing to try is to check the latest CVS version of LiveIce - the sample rate conversion code has changed a lot from the version of Icecast which is on icecast.org - specifically it removed a number of high end frequency artifacts. But Lame still doesn't compete well with the commercial encoders at low bitrates - although it is by far the best of the free encoders. Blade may make some claims about being able to compete at high bitrates, but it is seriously bad at low bitrates. Anyway - with LiveIce I try to put all the sample rate conversion code inside liveIce because some mp3encoders (l3enc) are much slower when you give them a 44100 kHz stream and expect them to output a 22050 kHz stream. And In the mixer mode it may need to perform different conversions for different mp3s. Now.... if Only I can figure out a way to integrate LAME without causing any patent problems... Scott Manley (aka Szyzyg) /------ _@/ Mail -----\ ___ _ _ __ __ _ | Armagh Observatory | / __| __ ___| |_| |_ | \/ |__ _ _ _ | |___ _ _ | Armagh | \__ \/ _/ _ \ _| _| | |\/| / _` | ' \| / -_) || | | Northern Ireland | |___/\__\___/\__|\__| |_| |_\__,_|_||_|_\___|\_, | | BT61 9DG. | http://star.arm.ac.uk/~spm/welcome.html |__/ \=====================/ -- MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
