Jake Hamby wrote:
> At the very least, I have a feeling that implementing a simple low-pass
> filter around 5kHz could remove the annoying high frequency artifacts I
> heard which was 99% of the problem with the LAME encoder for this task.
Yes, I find this too. By using the latest version of sox ("The Swiss Army Knife
of sound utilities"), I can put a sharp filter at around 5.3kHz, convert the
incoming audio to a 16kHz sampled .WAV, and use lame3.25(?) to get a nice
MPEG-2 layer III file with most of the `tinkling' in the treble removed.
In fact, you'll find that a recent audio file on the Linux Today website
(http://linuxtoday.com) was processed in just this way. It was the telephone
interview with the fellow who does Linux training certification, which the
web-site encoded at 32kbit/s using lame3.1-something at 32kHz sampling rate,
sent to me, I unpacked it with mpg123, processed it on my audio editor,
*re-encoded* it using lame3.25, and sent it back. Despite the *two*
encode-decode cycles, the new file sounds better than the original.
>From my point of view, as a sound engineer, musician and Free software
advocate, Lame is shaping up as an *incredibly* useful tool. If only I could
find a MPEG layer III FAQ to start learning about DSP properly.
John Hayward-Warburton
[EMAIL PROTECTED]
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
