I Built My Own Audio Player

I've been building a music collection in FLAC format for 25 years, and last year I bought an (Android) phone and a MicroSD card of 1TB that fits all of my music. It's been a long project for technology to catch up, but now that it's possible to have all of it in my pocket, I'm pretty happy with it.

I'm sure I can't be the only one that doesn't want to be a renter, give up control and stream anything the industry wants to push or deal with ads. It's cool to see some even go to great lengths to write their own application.

I've always had the issue on Android of the cover art/title being unreliable i.e. I change it and it just doesn't change, or it does then randomly changes to a random cover. As far as I could tell this was a bug in Android, did you run into this?

Technology has caught up many years ago, it's just that you insist on an format not fit for purpose. With good reencoding you get transparent audio quality (impossible to hear a difference) to fit all of your music on a much smaller card. (and as a backup you can always have those FLACs on the desktop)

You are very good collection curator! Only ~25% of my collection is FLAC/APE/ALAC/WavePack and still I have more than 3TB. It is what stops me from listening music on the go — I cannot choose what to put in mobile device in advance :-)

I have a Pi in my kitchen that I use as a Navidrome server. Of course, this only works so long as my phone has data reception.

I have also been building a personal collection of exclusively FLACs, be it for a lot less than 25 years. It's past 1TB but https://www.navidrome.org as the server and https://symfonium.app as the client has been great.

Granted, 2TB sd cards are now a thing so once they come down in price, I'll probably get one.

There are DAPs with two SD card slots such as [1]. There's now also 2TB sd cards, be it they're not cheap yet.

[1] - https://hifigo.com/products/hiby-rs2?variant=43134031167727

I've tried many media servers (but not Navidrome, though, I'll try it too!) and all for them mangle my collection. Albums shown twice (because there is FLAC and CUE files, for example), albums splitted into to tracks (each track seen as its own album, I don't know why), problems with non-unicode tags in old files, 6 ways to spell "Bjork", some don't understand FLAC with embedded CUE (and don't show tracks in such files at all), and, as a cherry on the top, none of them understand such abomination as ISO image with WavePack and CUE files inside (format which was popular on one big tracker some time ago). Files without tags are also a problem. So, each of these servers show 75% of collection Ok and 25% as hot mess. Each software has its own 25%, of course.

And I don't have any willpower to fix all tags and formats in 3.5TB collection (for example, to re-code all lossless zoo to FLAC and fix all MP3 tags for IDv2.4 format and Unicode).

Never seen Symfonium before going to have to give that a try looks great!.

I've been using Plex+PlexAmp for a while but have been really wanting to move to something outside of that.

I solved this by running a music stream. All songs i "like" are in a collection, and ices/icecast randomly select one song after another (i can also request things via discord bot) and i just fire up VLC and listen when i need music. Yeah its a little too random at times, but its also fun.

Even if we agree on a format I don't want someone to quietly say a song I like is gone from my library while I wasn't looking due to some reason.

> Technology has caught up many years ago

citation needed. Youtube still gives you crappy, unlistenable 153kbps crap.

There are a number of studies on this but this one has a good summary (1). The TL;DR is that over 256 kbps for MP3 there's no significant data that listeners could perceive a difference to CD quality audio. Lower than that you can perceive artifacts.

I'm too lazy for finding this but I recall this study or similar repeated for trained listeners (musicians and mastering engineers) with the same results.

Note that MP3 is 30 years old and newer perceptual audio codecs can beat it.

YouTube picking lower bitrates is a problem but the qualifier here is "at sufficient bandwidths."

(1) https://www.researchgate.net/publication/257068576_Subjectiv...

People should use wavpack for archiving instead of flac, to be quite honest. It feels like FLAC has mindshare and name recognition but it doesn't support hybrid encoding (which is great for storing audio for archival and playback) or more than 8 channels of audio.

Music has 2 channels, so that feature is of no use. The other feature is cool as it avoids the need t maintain 2 sets of tags, though as far as I understand, it's not widely supported, especially in smartphones

You're not limited by YouTube, it's your library, your encoding settings. And you don't even need any citations, do a proper AB test yourself to confirm the well established

Music absolutely uses more than two channels. Or less. You don't need special "support" either any more than flac, you just decode it and write the bytes to your output buffer.

How prevalent is music with >8 channels???

And it wasn't "special", it's just apps don't support playing the format

You don't pick an archival format for the common case, you pick it for all cases. But to answer the question, virtually every film soundtrack for the last 10 years.

I understand that pain. So many duplicate files, etc! Two or three laptops ago, I had a well-curated iTunes library, but now those same files are a tangled mess on the media server, as I'd overfill local storage on laptops & iPhones with the collection.

That doesn't make sense, you don't pick to suffer from poor support if you have a zero ore tiny niche case for it, you preserve it for that case only.

And you didn't answer the question, this discussion is about personal music libraries, like in the case of op collection of 25 years.

I'd bet almost all of them are stereo and not >8 channels

Perhaps. I've never gotten into the effort of re-coding everything, though I have converted some to MP3 to take on runs with me before memory cards exceeded the size of the data. I'm happy we're getting at a point not having to care anymore about disk space – that is, those of us not locked into a walled garden forcing expensive upgrades for more storage.

Does there exist an _objective_ method to test the sound quality difference between two files?

I could take a 54 kbs rip of Rust in Peace and reencode it at 256 kbs. It's not going to sound better than the 196 kbs rip, even though the bit rate is higher. What software would detect this? And other artifacts, such as clipping?

For some specific issues like clipping sure there are objective methods, and encoders use objective quality levels (so you don't target bitrate), and if it's not a tricky "subjective human experience can't be objective" question - the method to test is do A/B testing and see that humans can't detect a difference. That's audio transparent encoding. And since you don't care about other people for your personal phone use case, you can do this test yourself to detect the encoding quality level that's transparent to you.

And your examle doesn't make sense. You compare reencoded to the original, not to some 3rd sample. The issue here is whether to store large flacs or their smaller reencoded lossy variants. If you get a better quality flac, then you'll need to do another encoding to get a better lossy version

The idea is that once I have my CDs ripped to flac, I could run an encoder to MP3 for e.g. listening in my Tesla. But I don't know if the settings that work best for Dark Side of the Moon are going to be good for Rust in Peace. If there were some automated tool review hours and hours of audio, finding the discrepancies, that would save me a lot of time and provide me much enjoyment listening.