I haven't found a separate source for information on custom.conf and
custom-convert.conf but there is a great deal of useful stuff at the
start of the convert.conf file but I have much to learn. (It might be
useful to open that convert.conf in Notepad and copy the first section
into a separate file which you can then refer to.)
It is also worth looking at examples of custom-convert.conf entries
which various people have posted on the forum.
Much of the detail is explained in the convert.conf file there but here
are a few other bits and pieces:
Taking the Apple Lossless section in convert.conf:
Code:
--------------------
alc flc * *
# FT:{START=-j %s}U:{END=-e %u}D:{RESAMPLE=-r %d}
[faad] -q -w -f 1 $START$ $END$ $FILE$ | [sox] -q -t wav - -t flac -C 0
$RESAMPLE$ -
--------------------
The first line is explained in the information at the top of the file.
In the second line, the variables 'START', 'END' and 'RESAMPLE' are set,
so that when, for example, $START$ is used in the 3rd line it
substitutes '-j %s' where -j is the faad option and %s is the stream
seconds start offset (as explained at the top of the file).
The 3rd line defines how the transcoding is done. In this case it first
uses the utility faad.exe to decode the Apple Lossless to wave and then
'pipes' (or passes) the output of faad to sox.exe (that is what the "|"
signifies) which encodes the wave to flac and resamples where necessary
(dependent on the sampling rate of the file being played and whether the
player can handle it).
To see what faad and sox are doing you need to refer to the information
on both programmes that is easily available via Google. But be aware
that LMS uses versions of faad, sox, etc that are tweaked so there might
be some options not explained in the official documentation for those
utilities - however you can often work out what they are from the
context.
e.g. with faad (from faad documentation)
-q = Quiet - Suppresses status messages during processing
-w = Sets the processing output to be sent to the standard out
-f 1 = Set the output file format. (1 = Microsoft WAV format, 2 =
Raw PCM)
and with sox
-q = run in quiet mode (suppress messages)
-t wav = expect the input filetype to be in wave format
' - ' = accept the input from stdin instead of a filename
-t flac = set the output filetype as flac
-C 0 = set FLAC compression factor to '0'
You mentioned resampling and changing the bit depth, both of these can
be done with sox (which can also output in FLAC) but you need to read
the sox documentation and do further research, e.g. on the use of
dither, on using -v0.965 (say) to slightly reduce the volume to avoid
clipping, etc, etc .
As an example, this is an Inguz flac to flac that Phil Leigh posted some
time ago. Sox takes wave output from FLAC, upsamples it to 96000 before
passing its own output onto InguzDSP, which then passes its wave output
to FLAC to encode:
Code:
--------------------
flc flc transcode 00:04:20:22:01:8e
# FT:{START=--skip=%t}U:{END=--until=%v}D:{RESAMPLE=rate -v -s 96k}
[flac] -dcs $START$ $END$ -- $FILE$ | [sox] -q -v0.965 -t wav - -t wav
-C 0 - rate -v -s -M -a 96000 | [InguzDSP] -id "00:04:20:22:01:8e" -wav -wavo
-d 24 | [flac] -cs -0 --totally-silent -
--------------------
The sox 'rate' options that are used are -v (very high quality), -s
(steep filter (band-width = 99%), -M (minimum phase), -a (allow aliasing
above pass band) and the -v0.965 reduces the volume to 96.5%, presumably
to avoid clipping.
------------------------------------------------------------------------
JohnB's Profile: http://forums.slimdevices.com/member.php?userid=31553
View this thread: http://forums.slimdevices.com/showthread.php?t=96633
_______________________________________________
plugins mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/plugins
