It’s not every day that you see actually useful applications of AI, but this might be one.
It is probably good that OS community are exploring this however I’m not sure the technology is ready (or will ever be maybe) and it potentially undermines the labour intensive activity of producing high quality subtitling for accessibility.
I use them quite a lot and I’ve noticed they really struggle on key things like regional/national dialects, subject specific words and situations where context would allow improvement (e.g. a word invented solely in the universe of the media). So it’s probably managing 95% accuracy which is that danger zone where its good enough that no one checks it but bad enough that it can be really confusing if you are reliant on then. If we care about accessibility we need to care about it being high quality.
This is not by default bad thing, if it is something you only use when you decide to do so, when you don’t have other subtitles available tbh. I hate AI slop too but people just go to monkey brain rage mode when they read AI and stop processing any further information.
I’d still always prefer human translated subtitles if possible. However, right now I’m looking into translating entire book via LLM cause it would be only way to read that book, as it is not published in any language I speak. I speak English well enough, so I don’t really need subtitles, just like to have them on so I won’t miss anything.
For English language movies, I’d probably just watch them without subtitles if those were AI, as I don’t really need them, more like nice to have in case I miss something. For languages I don’t understand, it might be good, although I wager it will be quite bad for less common languages.
There’s a difference between LLM slop (“write me an article about foo”) and using an LLM for something that’s actually useful (“listen to the audio from this file and transcribe everything that sounds like human speech”).
Exactly. I know someone who is really smart and works in machine learning and when I listen to him in isolation, AI sounds like actually useful thing. Most people just are not smart like that, and most applications for AI are not very useful.
One of the things I often think is that AI makes it possible to do things that shouldn’t be done very easily and fast, that would had previously been too much effort or craft for some people, like now they can easily make website for whatever grift they are pushing.
While I hate the capitalist AI-apocalypse with a passion I think this is great news for accessibility.
Oh so that wasn’t a joke from their booth.
This seems really out of place, but locally ran auto subtitles from ethically sourced AI would be great.
It’s just that there’s two very big conditions in that sentence there.
Not against this feature, but this quote made me laugh:
… once this is in place, people won’t have to scour the internet for sourcing subtitles to their favorite movies, shows, or even anime.
As if MTL will get anywhere near the nuance of a properly made human translation.
Personally, I would be happy even if it didn’t translate it but were able to give some half decent transcription of, at least, English voice into English text. I prefer having subtitles, even when I speak the language, because it helps in noisy environments and/or when the characters mumble / have weird accents.
However, even that would likely be difficult with a lightweight model. Even big companies like Google often struggle with their autogenerated subtitles. When there’s some very context-specific terminology, or uncommon names, it fumbles. And adding translation to an already incorrect transcript multiplies the nonsense, even if the translation were technically correct.
deleted by creator
What do you mean by active component? Is processing the audio being played back to add subtitles active?
Sending the audio to an LLM in the sky. But I assume it would be local?
It says pretty explicitly that it only runs on the user’s machine.
deleted by creator
Even non-AI subtitles are off by default, what exactly are you expecting to be on?
deleted by creator
This is the Internet, there’s no shortage of targets.
The way you wrote this, I thought you meant that if it required a cloud service you would turn it off. But now I think you’re just saying you wouldn’t use this feature.
I share the confusion over your definition of “active”. You got all defensive when someone asked, so now no one really knows what you meant.
I’ve seen some pretty piss poor implementations on streaming apps but if anyone can get it right it’s VLC
If it’s opt in/opt out then am fine with that.
Not only is it opt in, it’s also running fully locally on your machine.
My biggest issue with that is the amount of bloat a full local LLM implementation would add.
But if it’s an optional module that you can choose to add (or choose not to add) after the fact, I have no complaint.
Ohh I assume it’s Mistral cause Llama uses a Incompatible license.
It’s not an LLM, just a subtitles generator for video.
It’s Whisper.
I wonder how powerful a device you need to run this live a la YouTube auto caption-style.
Does anyone have experience with this?
OHHH okay
Pandora’s Box is already open. Might as well make use of it.
Aaaaaand I drop VLC. Fucking shame.
Edit: “wtf i love ai now”- this thread
Why would you need to do that if it’s off by default and locally processed?
Because triggered and hate circlejerk.
Nuance is deader than Elvis.
uh huh-huh.
Cause we can no longer sit back and allow AI infiltration, AI indoctrination, AI subversion and the international AI conspiracy to sap and impurify all of our precious bodily fluids.
Is it off, or is it an optional module that doesn’t have to be adding bloat to my system if I don’t want to use it?
LLMs can take up a pretty big storage footprint.
Why don’t you ask them? They’re very responsive to their community of users.
I just took a spin through their news blog and changelog and didn’t see anything about it in the latest release, so it’s probably not out yet.
Do one thing and do it well. Oh well…
“Do one thing well” is what gives you software like
sendmail, which requires several other programs to be actually useful, all of which have to be configured separately to work together, with wildly different syntax.And enables modular workflows and flexiblity.
Meh, I’ll just stick with
mpv.How is MPVs impementation? Does it work fairly well?
Its a command line multimedia player. It’s implementation is ideal for minimalists, and easily understood by reading the man pages.
It works very well imo.
It won’t be better than human translated ones but begter than no subtitles. I don’t think even humans can make subtitles correctly without knowing context
Honestly, if it can generate subtitle files it’ll be a huge benefit to people creating subtitles. It’s way easier to start with bad subs and fix them than it is to write from scratch.
Yeah true. Good feature anyways
deleted by creator
