When I add a video release, MB creates a new recording automatically, which I change to video recording manually. But does this recording include audio as well (it is one whole, video + audio) or it include only video? If so, then I need to create separate audio recording for every video release I ever added, remove all sound-linked relationships and relink them to new recordings. But it needs too much forces… And it looks over-complicated for cases when only video release was publish (most of my video releases are those).
Here there are 4 possible options of adding video/audio releases based on publishing data and my current behavior for them:
Some audio recording was released, after that a video for it was released (most of cases in pop music IMO). In this case it is definitely 2 separate recordings. All relationships linked to audio should be added in audio recording, linked to video to video. The audio and video recordings should be linked with ‘music video for/has music video’
Video and audio recording were released at the same time. Same behavior as above
Video was released firstly and only after that the audio recording [from it/was used in it] was released. Usually I don’t know it was published in audio streaming platforms, but if I see that, I create a separate recording for the audio, and after that the above algorithm is used. But in this case release dates are a little bit strange, because MB doesn’t use/mentions the data, when the video using this audio was released (If some video was published in 2021 and the audio from it in 2022, then 2022 will be used as first release year).
Only video release was published. I link all relationships (sound/video) to the video recording, because in other case it is over-complicated. [Example].
So for me, while a video recording doesn’t have a ‘music video for’ relationship, it includes video and audio as well. I want to check, if my understanding of it is identical to other participants’ understanding and practicing.
in all those cases, you could add an audio recording to go with the video, as the audio would technically have different credits (basically just excluding artwork and other video credits). for case 4, I’d just create a standalone recording
that said, it’s of course not required, so you could skip it, it’s just a good practice in my opinion
As I said, most of my releases were published only in video format on YT and this is exactly what I did at the beginning after I learned from @chaban that video recordings should always be added separately from audio recordings. I took this practice based on chaban’s edits (here, for instance). As far as I can see, chaban prefers to add standalone recordings for video and not releases. But then here you advised to add releases for video recordings:
This is the main reason why I started to add releases for them. And now I have ~200 recordings like that one given in the example above.
This is so strange for me in MB. The main benefit of DBs is structuring of the data, but here there are so many fuzzy rules and recommendations, that in many cases you don’t have to follow them, if you don’t want XD
I’m not against to do that, but it is so frustrating that it can’t be made automatically. Every time after adding a video release, I need to open linked recording and to change checkbox to “video” manually. If audio even in video recordings should be added separately, then why if I add video release it can’t be generated automatically too (video + associated with it audio rec). It is not difficult to open in the end 2 tabs with video and audio and add rels. But if I need to add audio recording by me it is much more difficult, cause I need fill all duplicating data from the beginning. The main problem here is not difficulty, but the time I need to waste to do that… just for duplicating almost the same data before I can add sound-specific relationships. Really, it is so OVERCOMPLICATED. Sorry for showing my emotions so openly
I mean, it’s not a standardized practice, and it’s not in the docs. even some stuff in the docs is best practice sort of stuff, save for maybe the style guide
another part of it is we as a community are still figuring out how to handle videos in the database, so there isn’t much consensus nor official guidance on the subject
this is part of the reason there’s no automatic creation for such audio recordings
the main reasoning for that is if the audio hasn’t been released by itself, there wouldn’t be a release to attach the audio to. I’ve also done the same if an artist mentions a lost demo or even for one of my own unreleased tracks in order to log listens on ListenBrainz
I typically do this when I add such a video’s audio to my library, that way I can log listens to the technically correct recording
This is a very timely post for me, as I am trying to figure out many aspects of adding music videos, concert/festival videos (both bootleg and official), and movie/documentary videos with music content. For this reply I will attempt to stay within the original post, dealing with the video recordings and not ask about other aspects of adding a video release. So far the 5 video releases I have added (or worked on) are in “case 4”, “video only release”, and they come from DVD, not streaming service.
I have been ripping those DVDs multiple of times using MakeMKV, HandBrake and VLC in order to figure out the best process for me. I use the chapter time from video (using HandBrake) and I preserve the original video recording without transcoding (I think). The next step (which I have not implemented yet) is to extract the the music only part of the video container in its native encoding (if possible), and that gets us to “case 4”.
So this standalone recording does not generally exist, except for those that have extracted it in some way from the video (DVD or streaming). I wish to add all relationships (were available) and acousticID, is this permissible? It almost feels like its a “personal bootleg” which I think we are not supposed to add to MB. I am looking for guidance.
A second question around the video recording is to the intro’s, band introductions, outtakes, documentary discussion, etc, all of which may or may not include music but at least spoken words. Do these count as “video recordings”? (Maybe that is part of a different discussion that I need to create later on.)
I think if you pull the audio out of the video, it could be used for generating an AcoustID. I wouldn’t add it as a release, which I believe is the intent of the guideline
since a video file is a container (sorta like a .zip), it should be fairly easy to pull the audio file straight out of it with the right program
we allow most of these for audio and video releases already, as long as they’re music-related (tho I don’t know how many are adding documentaries not made by the band or their label). there’s several such documentaries on bonus DVDs of CD releases that I know of at least
I don’t know that I’d personally create standalone recordings for non-music, but I also see no issue with doing so, since we do also allow non-music (podcasts, audiobooks, interviews, etc.)
I still have more work to do and have no idea of the Blu-ray contents which should be the same as the DVD. It has commentary (spoken words) within some of the video chapters with the music very soft in the background. This is one of those I will start a new topic on some time in the future.
Thanks, after a lot of searching I have them, but am still learning the different uses. Handbrake has issues with not using libdvdcss-2.dll, plus the portable.ini is not read so I cannot stop the creation of the temp and log files on my SSD system disk. MakeMKV is pretty easy but it does not allow editing of the chapter names like Handbrake does. I am still looking at MKVToolNix for the audio extraction but would prefer ffmpeg for that. At some point I need to get a Blu-ray player. Lots to learn!
I know the reasons, but it is still so frustrating . My PC and internet connection are not fast, so for every additional operation, that I should do, I need to waste 3-5 seconds of my life, when I just wait and I can’t do anything else at the same time. For every video release, there are plenty of these actions, so in the end I waste 2-3 additional minutes that could potentially be saved (On average, I add a video release about every 10–15 minutes).
Sorry, but I don’t really understand what you exactly mean. Do you mean linking of listens with recordings on ListenBrainz? Because I just link my listens to suitable video recordings currently (you can check my LB profile), because I didn’t think before the writing this post it can be a problem. I believed video recording includes video and audio as well.
Based on questions of @dashv , I have a new one too. If I create an associated audio recording, what exactly should it have? All audio content that there is in video including intro/outro/some explanations by artist etc or only the track itself? What if they’re mixed as here, for instance? Nika Lenina speaks about something and her speech is nor part of music but it is overlaid on the track at the beginning. For me it is clear, that the associated audio recording always should include all audio content, because it will have other audio fingerprint and is separate from the usual audio recording that published on streaming services. But what do you and other participants think about it?
it’s more personal preference, I like to log my video listens seperately from my audio listens
honestly, I haven’t given this much thought, since I just do it for my library, and don’t submit AcoustIDs for these (tho maybe I ought to…), and usually don’t have to worry about a different fingerprint from streaming services since much of the music I listen to isn’t on the big ones (Spotify, Apple Music, Deezer, etc.)
to be clear, I personally only create a standalone audio recording when there isn’t already a non-video recording, and usually this is for lyric videos and visualizers that usually don’t have any added audio. beyond that, I don’t know that there is a right or wrong way to add these recordings
. My PC and internet connection are not fast, so for every additional operation, that I should do, I need to waste 3-5 seconds of my life, when I just wait and I can’t do anything else at the same time. For every video release, there are plenty of these actions, so in the end I waste 2-3 additional minutes that could potentially be saved (On average, I add a video release about every 10–15 minutes).