Nice work automating this process! I had a similar idea but music licensing is tough so I never thought it could survive in an app store without getting sued into oblivion.
Maybe if it only read the user's own library of songs?
Are you thinking of turning this into an app? I'm curious if you're planning to make this a business or if it's just for fun.
I want to make my fun webapp SuperAnimo[1] into a business, but so far I haven't found product-market fit.
thanks mrieck! currently is just for fun, if i got sued i will take it down and open source it.
SuperAnimo looks very nice (idea and implementation) but yeah, product-market fit is hard to achieve.
As a karaoke fan and singer I like it! I just tried a random song off the top of my head, and it was basically perfect. The voice and instrument were splitted perfectly and the captions were in sync (caption 2 was better; 1 was too short).
I agree as a consumer this isn't super useful even as a karaoke fan; just because every song you type in, it takes a while to work on it (understandably). And like GP said, licensing issues probably make it hard to market this for consumers. But I wonder if you could market it to all the companies that make and distribute karaoke videos for songs (to karaoke shops). I know nothing about that industry, and would not be surprised if there's already tooling that exists for this in the industry, but at least this might be a competing way to do this more cheaply.
Alternatively, there's a lot of Youtubers who cover pop songs. There are probably ways to work with them so they could easily locate vocal-free versions of the songs they want to do.
hi godot, thanks for your detailed feedback!
i really think about product&fun before money.
shorter processing time requires costly GPU..
as i see it, it's a really useful tool for the long tail, because for the popular songs you can use other great services thats produce it professionally (like sing king youtube channel), but the total count is only 30,000 songs from total of 60 millions songs!
Neat idea! Reporting my experience: I've tried with November rain but it said the song is too long. Probably it would be helpful to state that in the research bar (?).
Then I tried with welcome to the jungle, and the karaoke effect was very very good, but unfortunately the lyrics were totally unsynchronized, by 10 seconds or so. That is a pity as it completely undermines the purpose of the tool...
hi ghego1, thanks for your feedback.
lyrics to audio alignment is unsolved problem yet, so i use the best tools in existence, which give very good result for 50% of the songs in almost any language.
you can learn more about it here https://www.music-ir.org/mirex/wiki/2019:Automatic_Lyrics-to...
This is fantastic for an automated tool! It seems to work really well in the few cases I tried it with.
One area it could be improved in that that lyrics sometimes overlap. As far as I can tell, it always shows 2 lines of lyrics, but sometimes one goes over the other so that they're hard to read.This is probably just a spacing issue that could be fixed with some CSS.
they don't say how or where they get the info, and failed to get lyrics of a song, despite recognizing and listing it in the dropdown...
no mention of WHERE it tried to get the lyrics and failed...
hi jan6, youka here. i'm still in alpha stage, so bugs may happen. i'm search lyrics mainly in public sources. tell me which song you want, i'll fix it for you.
Maybe if it only read the user's own library of songs?
Are you thinking of turning this into an app? I'm curious if you're planning to make this a business or if it's just for fun.
I want to make my fun webapp SuperAnimo[1] into a business, but so far I haven't found product-market fit.
[1] https://www.superanimo.com