![]() ![]() |
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
70.166.116.8
I'm normalizing gains on my existing digital music library using Audacity. I'm currently doing a project for each album one at a time. There may be a batch process available but I'm finding audacity doesn't preserve metadata so it appears I'm going to have to do each LP one at a time. Ie. 90% of the effort is going to be in preserving tags. This becomes a job that'll take me from 9 to 18 months working an hour or two a day.But there may be better ways for me to go forward.
1.) Get Audacity to better handle the metadata for my project. Reducing having to fix any metadata is a big time saver for me. I'm losing album artist, discnumber and cover art. Audacity is changing track # (unless I'm careful and always creates single digit #'s instead of 01, 02,...,09) and renaming (song) Title to filename. It doesn't change filename but I'm using a discnumber+track# Title(song name) file naming convention and expect the metadata to match the filename.
2.) It would be faster if I could get Audacity to perform it's function on the whole library in a single step and create a shadow library with the hosed metadata for me to go fix before integrating the new files into my existing active project. The process of doing these one at a time manually increases the chances of errors. There are lots of steps, checking and double checking and a great potential for stupid errors that have to be fixed. There's 2700 albums and 1/2 half the time is spent making sure things aren't getting messed up and half that time is fixing them when they do. If there is a way get Audacity to do it's thing without my involvement it would be great.
3.) Suggestions?
4.) I wish Audacity would not change any metadata except whatever is required by it's processing. I could manually do an album in less than 1/2 time, probably much less. Maybe it does and I just don't see how to do it.
FWIW I've been using the amplify effect for a while when creating new albums for the digital library. But I need to equalize channel imbalances so I'm using loudness normalization instead. I like this as it assures reasonably consistent gains levels between all albums and "fixes" channel imbalances where gain adjustments weren't carefully made in the first place.
This makes a massive improvement in sound quality when listening to my digital music library and I'm going to spend the time to fix it. But if I can save time I'll be able to get around to taking photos of my collection and put them into the library and attach them to my database sooner.
Edits: 04/06/25
The first response to my comment actually got me thinking better about what I was trying to. I guess in the simplest context I want to produce lossless recordings of my analog LPs. I don't care about changing volume when playing vinyl why should I care when playing digital copies of vinyl. That said things like random play are better with volume equalization and I will look into doing things that may be applied at runtime without messing up the digital files (like upscaling or making fine channel balance adjustments).
Which leads me to getting better recordings of my LPs and that's upgrading off the 16 bit recorder.
IMHO this is the best and easiest way to record vinyl. Pricey but solid.
Maybe you could find the previous model used?
![]()
Cut to razor sounding violins
Best and easiest is a big claim. Personally I suspect but can't prove without actually comparing that any RME interface can produce better digital than any of the Tascams. That said if talking about free standing recorders I'm not sure the top of the line DA-3000SD actually produces better digital than the SSR250N and can't know without comparing. Sure it does a lot more but if someone isn't interested in doing more it's not worth spending the extra money unless saying they have it is the point. And it's not the easiest since the SS250 has an internet connection and there is not need to transfer files with physical media as required when using the SA3000D. But then that internet support on the SSR250N could actually affect small signal performance on the device. The problem with the RME interfaces is my PC has to be involved with the recording and to me that is potentially a clusterF of extra problems to deal with.
And a further complication with this is for me the idea of having my system DAC, ADC and headphone amp in a single box is somewhat contrary to my basic audio philosophy. All that said the Tascam DA-3000SD is either the best choice or the worst depending how I am thinking about things at any particular minute. I might buy it.
Yeah nothing is perfect. FWIW, not having to split files is a huge benefit IMHO. Maybe you have a way around that, but when I tried other methods, it was just a pain to have to split the files in Audacity. With the tascam it just does it if you set the levels right. Sure some albums are tough if they don't have silences but for the most part it just works.
And you can record off the web, like youtube and its just put in the sd card hit record then play on the computer and viola.
Also you CAN play the files back with no need of a computer and as you mention recording too needs the computer with other solutions. Also not sure if DSD is a big deal for you but the DA-3000 can record at DSD resolution.
Not sure about the SSR250n. It may be leaps and bounds better but no clue.
Also not tying in the audio system with the computer might have an improvement just because the grounding is better. And or you can do some computer work while recording and not worry about affecting the sq.
FWIW the headphone jack is pretty good on the tascam but you could always run it out to a dedicated amp. I do both myself depending. The main amp has some processing, soundstage expansion, BBE Sonic Maximizer, crossfeed bass boost. The tascam can give a direct unprocessed feed if I like.
These days the tascam is my main dac/amp for computer duties and might make it in the main system if anyone could help me use it as a DSD dac, but that seems an impossibility...
![]()
Cut to razor sounding violins
The Tascam DA-3000SD is the best stand alone AD recorder I've ever considered. Enjoy it has much more to offer than the 16 bit recorders I've been using for the last 15 years. Thanks for your inputs - the dbPoweramp stuff was particularly useful but all comments are appreciated.
When you fill in Tag info during the Export, save the tag! Then you can Load that Tag file as needed in the future and update the tag before Export.You can also Edit the tag XML files, to update the Comments for example. I use Comments to store the equipment used for recording.
My problem with recording LPs is that every couple of years my vinyl rig improves. I struggle with a perceived need to record the LPs again.
I record to 32/384 because it sounds better. Each album side is about 4 GB.
Edits: 04/29/25
Thanks. Seems to work as you describe. I'm going to bail on this project - too much work. Probably just rerecord the worst of them can pick up a couple of bits of resolution on those. These sound pretty good but I bet with proper gain settings they'll be even better.
I'm interested in a new recorder. How are you saving digital with those RME ad/da devices? Laptop and what software? The Black Edition looks interesting. The connections are confusing to me. Analog input to record would come from tape outs on pre but don't see rca inputs on the RME. Looks like I could probably select a digit output for the converted signal and a digital input so I could use the RME as the DAC for my audio system with the output going to my preamp via analog outs. I've never used an optical connector but that looks like the only way for me to get digital into the DAC.
Regarding gain for recording, the goal is to set gain as high as you can without clipping. You can see clipping in Audacity. I use an RME ADI-2 Pro FSR Black AD converter, set to Preamp mode, to set gain.
You are correct that there are no RCA inputs on the RME. An adapter can be used.
I use a USB cable to connect the RME to my computer, and a USB cable to connect the computer to my playback device, a Luxman DA-07x.
I agree that it is a lot of trouble at first, but one conquers the learning curve fairly quickly.
One more thing how are you saving the digital from the analog input to your RME via the USB out to your laptop? What SW are you using?
Thanks in advance
The SW is Audacity. The music is *Exported* with the preferred file name, file type (WAV, FLAC, etc.), and bit depth. The sampling rate is selected in Audacity. I save as WAV, 32-bit 384kHz because it sounds better.
It seems a dedicated laptop or workstation is required?
Dedicated? No. Something to install Audacity on? Yes.
And would you expect modifications to OS priorities or other running applications to insure a PCs ability to serve as a recorder?
I probably should figure out a way to test my computer if I'm going to spring for either of these interfaces.
In Windows, disabling unnecessary Services and Start-up apps is usually recommended.
Maybe someone will post of list of these for Windows 11.
Ran some loopback captures with Audacity yesterday and the recordings came off just fine. I'll be doing more today. No problems. Will expand the test later to actually record off an external device.
Thanks for your help. The RME Black Edition looks great and using it as a headphone amp and a system DAC is a double bonus.
My current Tascam stores the converted digital directly to a CDRW and higher resolution Tascam provides ssd drives on the recorder.
Been a while but doesn't dbpoweramp have a batch converter? Bet there is an option to manage gain
![]()
Cut to razor sounding violins
I use dbPoweramp as the last step in my current process. I take completely tagged digital LPs, usually 20-50 at a time, and dbpoweramp renames them and places them in a working directory tree album_artist-> album-> tracks. The naming convention is discnumber+track# title(name of song). Then I manually copy them over to the music library. I do the last step manually so I don't end up copying a duplicate album name into the artists directory - when that happens songs will get copied over.
I took a quick look at dBpoweramp and yes it has some features I am looking for. I have to look at those to see if they are going to be useful. The big problem is going to be keeping album song volumes relative to each other - which is kind of what makes it hard using Audacity.
Going forward I'll be doing this kind of processing in Audacity before I break the sides into tracks. But for cleaning up the existing 2700 album library the dBpoweramp batch processer could be a massive time saver.
Thanks for bringing that up.
Awesome. Hope it works. They might have some way to make it relative even if it isnt. Like capping at a certain volume? I recall so many options there it was crazy. Bet you do well.
![]()
Cut to razor sounding violins
I'm no Audacity expert but I think it's great for tinkering as I did when I digitized a dozen albums then decided it wasn't worth the hassle. It was a learning experience and fun! I used it mostly for splitting tracks, reducing objectionably loud pops and clicks but I left minor noise alone. I had to manually split tracks and manually input metadata for each track. Man what a chore!Specialized software for digitizing vinyl can help automate the entire process but I don't recall what they are. A couple guys here in the Asylum have gone that route.
Edits: 04/07/25
I only used Audacity in the processing of LP sides recorded to CDRW. It's the best way to split the side into songs and name (initially) the tracks. I update tags using mp3tag. Rename and create artist/album/tracks directory using dbpoweramp. Then manually, with file explorer merge those directories into the music library. Then use BeyondCompare to update 2 backup drives. Then wipe the hd used in the audio system clean and copy a fresh version of the library to it.
But in the last few years of been integrating some of the Audacity effects into the processing and want to go back and add them to previously created work.
It is a part time job.
Wonder why you use Audacity. It is an audio editor but if it is about loudness normalization, any media player can do this and will simply write a tag. No need to edit (and dither) the samples.
Wonder why you are losing tags. Are you using WAV?
The Well Tempered Computer
When I started to use Audacity it was exclusively used for breaking album sides (it's a PITA doing it on the recorder). When splitting I'm also entering the song title and track number and the file is saved into a temporary working directory album_artist/album/tracks where I then update tagging with mp3tag. I might have 10-60 records in their own working directory when I use DBpoweramp and create a temporary directory structure consisting of album artists/album/tracks with tracks named as "discnumber/track# title". Then I manually merge these temporary artist directories in the music library.I'm converting wav to FLAC when I rip to the computer. It is about normalizing volume across all albums - it's distracting to have to adjust volume levels from album to album and even more so from song to song if using random play mode.
But it's also to correct for unequal channel gains at recording time.
-------
Note : This project start over 15 years ago when I was recording lossless music to play at the hockey rink when I was acting as a scorekeeper for my son's team. Kind of evolved into getting my favorite albums so I take them on portables then into getting all my vinyl into a digital format (took 4-5 years) to having a HD as source in my main audio system about 5 years ago. And since then I've been updating as I acquire new albums (50-120 a year). I started using amplify to increase levels on recordings where I set gains too low but find that less distracting than when I did not set the gains equally so I'm now using Loudness Normalization to adjust that.I've been very happy with the results and am now going back over the existing library and updating it accordingly. FWIW the digital music library (flac only) has been sourced from ~2000 lps, ~400 cds and almost 300 lossless live or broadcast concert downloads (flac).
Are you saying these music files will/may sound audibly worse after this processing?
Edits: 04/07/25
This post is made possible by the generous support of people like you and our sponsors: