Dear all, I am an active FLAC user and like the program a lot! Thanks for your great work. Since I am not a developer I cannot modify the program on my own. Maybe you might be interested in features I would find useful. Maybe others find them interesting as well and you find the time to implement them. 1. Add encoder options: a) Skip silence (leading and ending) would give the option to skip silence at the beginning and end of WAV-File. Currently I have to use an external program (WavTrim) to do the job. b) Convert to mono: I encode a lot of audio books from CD. While in reality they are only mono (one person reading) the are transferred as stereo from the CD. Converting them to mono greatly increases overall compression. Currently I have to use SoX for conversion which is slow and painfull. Better would be to have an option in FLAC to skip one channel and encode the other channel as mono (as seen in LAME) 2. Develop DLL for EAC I use Exact Audio Copy (EAC) for CD ripping. It would be great to have an DLL (as in Monkeysaudio) to directly rip to FLAC and not have a WAV-file as intermediate 3. Album-gain based on individual replay-gain values I encode each file with an replay-gain value. If I want to add album-gain values with metaflac all files are analyzed again. Could the album gain value be calculated by using existing track gain values for the individual files. This operation would be much faster. Do you find this suggestions interesting? Regards, Alexander -- Lust, ein paar Euro nebenbei zu verdienen? Ohne Kosten, ohne Risiko! Satte Provisionen für GMX Partner: http://www.gmx.net/de/go/partner
--- Alexander <alex1452@gmx.net> wrote:> 1. Add encoder options: > a) Skip silence (leading and ending) would give the option to skip > silence > at the beginning and end of WAV-File. Currently I have to use an > external > program (WavTrim) to do the job. > b) Convert to mono: I encode a lot of audio books from CD. While in > reality > they are only mono (one person reading) the are transferred as stereo > from > the CD. Converting them to mono greatly increases overall > compression. > Currently I have to use SoX for conversion which is slow and > painfull. > Better would be to have an option in FLAC to skip one channel and > encode the > other channel as mono (as seen in LAME)both of these are non-lossless things that really don't belong in flac.> 2. Develop DLL for EAC > I use Exact Audio Copy (EAC) for CD ripping. It would be great to > have an > DLL (as in Monkeysaudio) to directly rip to FLAC and not have a > WAV-file as > intermediateEAC is closed source so its author has to be the one to do it. he did the monkey's audio one.> 3. Album-gain based on individual replay-gain values > I encode each file with an replay-gain value. If I want to add > album-gain > values with metaflac all files are analyzed again. Could the album > gain > value be calculated by using existing track gain values for the > individual > files. This operation would be much faster.album gain can't be calculated from the individual track gains, the whole album has to be analyzed together. Josh __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com
I vote for #2. I don't see the options in #1 as strictly going with the philosophy of FLAC being a lossless codec although neither one seems like a bad idea if implemented properly. More generally part b of #1 would allow you to select any channel(s) from a source and just encode those channels. For example you may want to encode only 3 channels of a 4 channel source or 5 of 8 or 1 of 8, etc. This seems like it would be the best way to do things, given what I know of the current FLAC architecture. I started fiddling with trying to learn some FLAC development a year or 2 ago, but unfortunately I have never really found the time and haven't touched it since. Between school and work I already don't get out much or I might be interested in doing #2. Alas, maybe when I finish grad school in a couple years if nobody has gotten around to it. Is there a proper etiquette for posting to this list as far as top post, bottom post, partial quote, etc? Devon -----Original Message----- From: flac-bounces@xiph.org [mailto:flac-bounces@xiph.org] On Behalf Of Alexander Sent: Monday, November 07, 2005 1:38 AM To: flac@xiph.org Subject: [Flac] Feature request FLAC Dear all, I am an active FLAC user and like the program a lot! Thanks for your great work. Since I am not a developer I cannot modify the program on my own. Maybe you might be interested in features I would find useful. Maybe others find them interesting as well and you find the time to implement them. 1. Add encoder options: a) Skip silence (leading and ending) would give the option to skip silence at the beginning and end of WAV-File. Currently I have to use an external program (WavTrim) to do the job. b) Convert to mono: I encode a lot of audio books from CD. While in reality they are only mono (one person reading) the are transferred as stereo from the CD. Converting them to mono greatly increases overall compression. Currently I have to use SoX for conversion which is slow and painfull. Better would be to have an option in FLAC to skip one channel and encode the other channel as mono (as seen in LAME) 2. Develop DLL for EAC I use Exact Audio Copy (EAC) for CD ripping. It would be great to have an DLL (as in Monkeysaudio) to directly rip to FLAC and not have a WAV-file as intermediate 3. Album-gain based on individual replay-gain values I encode each file with an replay-gain value. If I want to add album-gain values with metaflac all files are analyzed again. Could the album gain value be calculated by using existing track gain values for the individual files. This operation would be much faster. Do you find this suggestions interesting? Regards, Alexander -- Lust, ein paar Euro nebenbei zu verdienen? Ohne Kosten, ohne Risiko! Satte Provisionen f|r GMX Partner: http://www.gmx.net/de/go/partner _______________________________________________ Flac mailing list Flac@xiph.org http://lists.xiph.org/mailman/listinfo/flac
Another option for 1b is to have an option to detect and discard empty channels. That way, header information could be lost, but audio data could not. This option would require reading the entire empty channel before encoding. OTOH, this is arguably a ripping issue, not a compression one. And besides, shouldn't an empty channel compress to negligible size? 1a seems clearly lossy. There's usually a reason for a period of silence (as opposed to an entire channel that's always silent). IMHO. -- Brady Patterson (brady@spaceship.com) RLRR LRLL RLLR LRRL RRLR LLRL On Fri, 9 Dec 2005, Devon Holcombe wrote:> I don't see the options in #1 as strictly going with the philosophy of FLAC > being a lossless codec although neither one seems like a bad idea if > implemented properly. More generally part b of #1 would allow you to select > any channel(s) from a source and just encode those channels. For example you > may want to encode only 3 channels of a 4 channel source or 5 of 8 or 1 of > 8, etc. This seems like it would be the best way to do things, given what I > know of the current FLAC architecture. > > -----Original Message----- > From: flac-bounces@xiph.org [mailto:flac-bounces@xiph.org] On Behalf Of > Alexander > Sent: Monday, November 07, 2005 1:38 AM > To: flac@xiph.org > Subject: [Flac] Feature request FLAC > > 1. Add encoder options: > a) Skip silence (leading and ending) would give the option to skip silence > at the beginning and end of WAV-File. Currently I have to use an external > program (WavTrim) to do the job. > b) Convert to mono: I encode a lot of audio books from CD. While in reality > they are only mono (one person reading) the are transferred as stereo from > the CD. Converting them to mono greatly increases overall compression. > Currently I have to use SoX for conversion which is slow and painfull. > Better would be to have an option in FLAC to skip one channel and encode the > other channel as mono (as seen in LAME)
Alexander wrote:>Maybe you might be interested in features I would find useful. Maybe others >find them interesting as well and you find the time to implement them. > >The only thing currently I would like to see is some sort of summary after encoding/decoding/testing. Would make it easier to see if the rip came out okay... The fact that at least on my PC, the CMD cuts off at about 35 tracks ripped, it would be down right useful! :D Sean