Subject: Re: [linux-audio-user] ecasound problems
From: Kai Vehmanen (kai.vehmanen_AT_wakkanet.fi)
Date: Thu Dec 06 2001 - 07:02:45 EET
Answering to an older post...
On Mon, 19 Nov 2001, dave willis wrote:
> however, i cannot select these options for oggenc (or lame) on the command
> line, but only in ~/.ecasoundrc, and i use different sample rates and
> channels. if you could consider adding this functionality into ecasound
> (ie: an option to pass options), it would be great.
I've just committed a few improvements related to ogg/mp3 output to
ecasound's CVS-tree. You can now for instance:
ecasound -i foo.wav -f:16,4,22050 -o test.ogg,192000
The above encodes 'foo.wav' to ogg, where the resulting ogg file is
16bit, 4ch, 22050Hz, and encoded with bitrate of 192kbps. Format of
'foo.wav' doesn't matter. If 'foo.wav' has less than 4 channels, then the
additional channels will be silent in 'test.ogg'.
It's also possible to encode in parallel to multiple mp3/ogg files, with
separate bitrate (mp3+ogg) and ch/srate/bitrate (ogg-only) settings. For
instance:
ecasound -a:1,2,3,4 -i /dev/dsp \
-a:1 -o test.mp3,128000 \
-a:2 -o test2.mp3,192000 \
-a:3 -o test.ogg,128000 \
-a:4 -o test2.ogg,192000
The above records audio from /dev/dsp and encodes it in realtime to four
different mp3/ogg formats. Ecasound's disk-i/o subsystem takes care of
buffering, so unless you really run out of cpu-power, it's unlikely that
encoder processes can cause dropouts to recording. Note that with these
newer ecasound 2.1devx releases it's no longer necessary to add options
like -z:db and -r to ensure reliable operation.
PS Currently you need ecasound's CVS-version to use the above. First
tar-ball release with these changes included will be 2.1dev6.
-- http://www.eca.cx Audio software for Linux!
This archive was generated by hypermail 2b28 : Thu Dec 06 2001 - 06:58:16 EET