Title: Organizing Ripped CDs and Renaming Tracks in Roon (ref#KWQSL6)

Affected Product


Roon Issue Category

Interface & Usability

Description of Issue

I ripped a box set using Roon Server - that is connecting a CD drive to the server and stuck the CDs, one by one. When it was done, turns out that Roon recognized just 2 of the 5 CDs in the box.
I used the “Identify Album” feature and while it was not a walk in the park, finally, ended up with correctly identified entry in Roon, with tracks properly aligned, etc. BUT

On disk the tracks are still the mess that Roon ripped - separate directories, generically named because the CD was not recognized, same for track files.

All this intro, to ask my question. I could not find (and I apologize if it is easy) a way to make Roon reorganize the files on disk? That is, consolidate into a single directory, given that they are mapped to a single entry/album and rename the files based on the track name/description?

Is there such a thing in Roon? I know that is available in other audio programs, so I assume it is in Roon and I couldn’t find it?

Roon Server Platform

Linux (NAS/SonicTransporter/Antipodes/Ubuntu/etc.)

Linux Server Type


Roon Server Specifications

Not a hardware issue

Connected Audio Devices

Schiit thru USB

Home Network Details

Cisco Business

Roon never changes your files and this is part of it

Ripping via Roon server does what it does and it’s really only for people who don’t care about folder structure or file tags. (Or messed up box sets). If you want control, use a ripper like dBPoweramp - it won’t get box sets automatically right, either, but at least you can make changes.


Following the conventions given in this Help article is your best bet…

But you have to do it, Roon won’t do it for you.


OOOHH - Got it

Well it makes sense if your only interface to that library on disk is Roon - once you adjust the identification, and the entry is good, the mess is hidden from you.

For some reason thought that Roon could rename the files based on the identification - some taggers do; unfortunately none that I know identify that boxset as well as the combo of Roon identification + my adjustments. I was hoping that work could be exploited to have ALSO at least file names based on track names.

Thanks for the help.

1 Like


So - after ripping - really look to invest in a good tagger (Yate is far and away the best for macOS). It will save you hours and hours; and is really the best way to go… IMHO :slight_smile: .

1 Like

Done some work comparing SongKong versus other taggers lke Yate here. Not totally objective but have tried to be fair.


Thanks for that!

Would you be open to feedback here, please?

I wonder whether Roon users would welcome another rounded and perhaps more complete overview of tagging software - in the case of at least one other product which you include, and that does very well by Roon.

This may reduce the chance(s) of anyone falling into the trap of assuming that SongKong’s is the default feature set; or indeed - by implication (twice as many features as has the next highest) a preferable one.

Not wishing to be combative, but additive, Paul :slight_smile: .

It would be great if there was a complete and independent tagging comparison and maybe this forum is the place to do it. But it certainly doesnt exist at the moment, the reviews out there are very limited and in many cases the reviewer doesn’t appear to have spent more than 10 minutes on the software, that was one of the main drivers for me doing this. In contrast for the reviews I have done so far I spent a whole day on each piece of software

And I am always open to feedback and discussion!

Regarding the feature set I have not listed every single feature that every single tagger application has so, but I have listed what I believe to be the most important features, and SongKong does have alot of features. The main one that SongKong is currently lacking is automated manual editing but im working on that.

Also, that is the summary list on that page if you select a tagger you get a more detailed list

Thanks. Then, may I offer a more fully-featured comparison with Yate, which I’ve used extensively for many years with my exclusively-‘Classical’ music collection in Roon?

The main reason for this is to point out that Yate has a myriad of features (not to mention outstanding technical support), which - respectfully, Paul - I don’t believe all emerge - even from your detailed comparison.

My intention is not to ‘pitch’ Yate: no one tagger is likely to suit everyone: thorough and appropriate tagging is a complex area of audio and music. Thankfully, Roon has unambiguous guidelines on the subject. Yate is both scrupulous and flexible in following them.

I’ve found it useful (and important) to understand the underlying principles on which Yate is based. My experience (having tried the other taggers referred to) is that these basic design objectives do distinguish Yate from the rest.

What’s more, to understand such distinction(s) benefits Roon users (on macOS) looking to become expert (competent, even :slight_smile: ) with tagging.

Yate relies less on the what some might call the elusive shibboleth of matching at all costs.

Rather, Yate puts the user in full control - perhaps most obviously and usefully with its amazing number of ‘Actions’ - a sophisticated scripting mechanism.

Sample Actions are only the start… Yate’s developer seems never to sleep and is always willing to help users with specific needs to achieve them via this inbuilt scripting etc which works so well in Yate.

I’d go as far as to say that pretty much anything can be achieved with these automation and (text) manipulation (etc) tools in what - for me, a long time user as I say - is in fact a highly intuitive UI.

Yate’s documentation is superb. In fact for the price at which Yate is made available, the range of features and functionality, the support and the robustness of the app are truly remarkable. Perhaps that is one of the reasons why it has such a strong following amongst users.

Yate is massively customizable - one of its greatest strengths; I’ve always felt in total control of using a very open structure (including spreadsheet-style editing, actually) to achieve equally extensive automation.

Yate is also actually very good indeed for Classical - again, if for no other reason than because of its open structure and huge number of ways of doing so many things to and with (single and combinations of) tags.

Yate integrates well with other applications… it’s ‘Open With’ functionality; it does support video files; and batch processing; and has a REGEX (Regular Expression) tester. Yate can read and write JSON and plist files; has ReplayGain calculation and integrates with Apple Music (service), iTunes, Music, TV integration; Yate has Spotify playlist creation; and Beatport as well as IMDb integration. In fact artwork can be resized ‘in situ’ and such formats converted. Yate supports all metadata in all audio formats.

So for anyone contemplating a tagger for use prior to import in Roon (the substance of this thread), your table is a great start, Paul - and thanks again; but could well be expanded, I think you’ll agree :slight_smile: .

1 Like

My recommendation is to tag all your rips with MusicBrainz Picard. It can be configured to do all the file naming and folder structures as well. There’s a bit of a learning curve but the greatest value is that Roon uses the MusicBrainz database (among other data sources).

If it’s tagged with MusicBrainz metadata, the likelihood is that it will match nicely when Roon scans it. If it’s missing from MusicBrainz or something is incorrect, you can make contributions or corrections.

Are the design objectives defined somewhere ?

But in my review I did say that

It takes a swiss army knife approach, in that it has many features but we did not find it very clear how they work together, or in some cases how they work.

So yes Yate certainly has alot of features and gives the user the ability to do all kinds of things and allow them to spend time being expert at tagging.

But it is my contention that most users do not want to become expert at tagging, they just want their music tagged correctly so it works well (with Roon). So the SongKong philosophy has been to do automated tagging at a level of accuracy and quality that it can be trusted so music can be tagged with the minimum effort, and provide options required to resolve certain situations and user requirements without providing options that could actually damage users music collections. If an album can be accurately matched online there is less need for the automated manual editing which is Yates main strength, having said that I am going to beef up SongKong automated manual editing because automated matching of 100% of collection is currenlty unlikley.

I think SongKong is currently the only music tagger that can batch match many albums in one go with the confidence it will not do a bad match, this cannot be achieved by Metdatatics, Yate, Beets or Picard for various reasons.

I found it a hard read for example this page on exceptions lists for case transformations

Can it do any of the following options, it wasn’t clear to me it could

ReplayGain is a good feature
AppleMusic is useful (SongKong also has this and in feature list)
Artwork Resize (SongKong also has this)

But I don’t think the other features you listed are revelent to a music tagger, they are just bloat.

So I could expand it further to include things like ReplayGain but as the list is always comparing SongKong with other taggers on my site it seems slightly odd to add many features that SongKong doesn’t have. I would love it if the list could be expanded and published by an independent source that we could interact with though.

I haven’t tried SongKong, but MusicBrainz Picard can accurately batch match many albums when you cluster them.


Thanks for continuing the debate this way :slight_smile:

The 2ManyRobots home page for Yate makes it clear that - as you say - Swiss Army Knives - will work well for many people: superb automation and an emphasis on customization that puts the user in control.

I have no means of knowing what ‘most’ users want. But - as both you and Barry would probably agree - there are many kinds of users with many kinds of needs. SongKong meets many; Yate also meets many. Although, from experience, I still respectfully submit that Yate (and others in your table, for all I know) does more than your comparison suggests.

My concern in posting as I did remains that users with no experience of tagging benefit from being aware of and understanding all the options; and that Yate really does represent a very rich mine of possibilities.

There’s a lot there, isn’t there, for sure. Once you’ve used Yate, though, it really does become second nature, I promise :slight_smile: .

== snip ==

Yes - with Actions.

There we shall have to friendlily disagree, Paul. There are many which I - and other users (here) - do use regularly.

Maybe it is - as you imply - a case of horses for horses.

Who could not be encouraged by your commitment to continued updating and enhancement! OTOH, I find Yate’s almost incredible richness of features (I still sometimes stop in the middle of an operation, wonder whether a certain function is possible, look in Help; and find it is!) extremely useful and always 100% reliable… no crashes etc.

For sure. Though I think it’d best be done by those who use each solution - with all their differences - regularly.

I’ve decided to post a replay although it is pretty much outside my typical behaviour. There are a lot of taggers out there. Use the tagger that does what you want and/or the one that you like to use. Most taggers have evaluation periods. Play with the product and see if it meets your needs.

While feature charts have their uses, they tend to represent the features that the author feels is important or reflects a single product. I can’t speak for other taggers, but there are certainly a large number of inaccuracies and omissions in the linked chart regarding Yate.

The only reason I’m replying is that I do take some umbrage to the bloat comment. I use many of my bloat features such as Album and Track databases every day. I certainly do not feel qualified to make decisions as to what features are core and absolutely necessary for a tagger. There are some taggers out there that do nothing other than simple editing. I have some users who are totally involved with iTunes/Music and have a complete different desired feature set than most of my Roon users have.

I hope I do not come across as confrontational. It is not my desire. I wish all small independent software developers the best. It may be a strange comment but I really do not care what tagger or player you use. Do what you will with your music and tag it as your see fit. Listen to the music!

As many of my users have been sending me emails about this thread, I’ve posted about it on my blog at 2manyrobots.com


I’ll add my two pennies worth. MP3Tag is available as a native app for OSX https://mp3tag.app/ . Yes it’s paid for but getting the metadata and naming right in your music files from the off makes it far easier for Roon to identify it.

The app is being updated regularly as well.


Hi Barry, yes the list is subjective but it is also my honest attempt to be accurate, So please if there are inaccuracies let me know the details and I will immediately correct therm.

Regarding omissions it was not the intention to list every single feature and option that every application has, we don’t even list all the SongKong features. But if there was a way to create an list in an independent location I would support that. One of the reasons for creating this list was that existing comparisons were superficial, for example if an application supported a MusicBraizn lookup it would be described as a MusicBrainz enabled tagger regardless of how simple, reliable, effective that feature was.

So my comment about bloat was regarding the features that Mark listed, and Album and Track database was not one of them. But I don’t think that an audio tagger needs to support video files or writing plist files. It does not mean these features are not useful for anyone, just that they are probably not useful for most users and in my opinion having them makes the application more difficult to use. A good example is ITunes, it started out simple enough and got more and more complex, then eventually they had the good sense to split into Apple Music, Apple Tv, Apple Podcasts. I wasn’t meaning to offend perhaps feature creep is a better term

1 Like

Hi Mark, so the first option in that list was the automatic identifcation of classical albums to non-classical albums so that the subsequent options can be applied only classical albums. SongKong has quite a sophisticated algorithm to do this, necessary because MusicBrainz itself has no flag to distinguish such albums. You say this can be done by Yate with Actions but I could not find such an option could you give me more details on how to do that please.

Hi Mike, Picard can can batch tag multiple albums but not that many before it struggles. It isn’t fully automated, after clustering you are meant to check each cluster before preceding to lookup stage, and then check each result. It doesn’t combine the benefits of acoustid and metadata matching the user has to decide one which one to use.

If the files are not organized into albums you are warned to not to process more than 200 files at a time


In contrast SongKong can be pointed at 10,000 albums and process them in one go. There should not be any bad matches but if there are you can revert individual albums to how they were beforehand.

Comparison with Picard here

Hi Paul,

I appreciate the reply. Yate does not support WMA as listed but it does support DFF, APE and WavPack. Personally I think APE tags are evil but I implemented them because I had a lot of requests for the functionality. (Largely from people who wanted to migrate the tags to other audio types). The list of fields which are missing can be implemented as one of the 100 available custom fields which in a lot of cases are automatically written when importing from MusicBrainz, Discogs, Beatport, and others. In fact my setup has Live, Single and Instrumental fields which I use to filter/search on in my databases.

Regarding classical extraction, there is an action statement named Decompose Title into Classical Metadata and a supplied action named Classical Metadata from Title. Both are misnomers as any reasonable field can be used as the source.

In my opinion it is somewhat pointless to go into more detail on omissions or inaccuracies as that would entail the production of an entirely new chart and … that’s simply not my thing.

As far as feature creep goes it all comes down again to who decides what is necessary and what is not. Being able to match 10,000 songs in a single go is a good feature … for some and perhaps for many. What I’ve found is that not everyone is concerned with having their metadata be an exact match of what is available on a single data source. Personally for various reasons I purchase a lot of music from Apple. Using the Apple Music APIs (not the application) I can extract metadata and artwork as represented in Apple’s enormous database. I also personally use MusicBrainz and Discogs in my never ending search for musician credits. Different strokes for different folks.

BTW: representing iTunes being split into Music, TV and Podcasts as an example of solving feature creep is not a great example. The sum of the three newer products still has less functionality than the last version of iTunes … at least as far as a tagger is concerned.

I wish you success, as I do the authors of all the other taggers represented in your chart.


I guess it depends a bit on your definition of “many”, then. :slight_smile: It’s not an import part of my workflow but I can see situations where people want to quickly organize a large library and it looks like SongKong would be a better tool for that.

I only rip/tag one album at a time and do a much deeper dive to identify the correct release and submit/correct metadata to MusicBrainz (over 30K edits so far). I highly doubt any automated process can differentiate between two pressings from different countries or pressing plants when they share the same DiscID and AcoustIDs. Sometimes the only way to identify the release is by the artwork or the Matrix.

I understand most people wouldn’t care as long as the song title and artist credits are correct.