Considering the progress on speech recognition, why do YouTube subtitles have little in common with what's spoken in the video? Shouldn't Google be able to do a good job at auch a task?

▲

Jabrov 3 hours ago | parent | next [-]

Scale is a big part of it.

Every minute, 500 hours of video are uploaded to YouTube.

Even Google can't afford to run the biggest baddest model on every single one of those hours.

▲

CodesInChaos 3 hours ago | parent | next [-]

Though they do have the money to generate terrible TTS audio for many videos. And since google can't fathom a person knowing a foreign language, use those automatically without warning.

	▲	lambda 2 hours ago \| parent [-]
		I've even seen videos that were in English, but mislabeled, so they got auto-dubbed back to English, replacing the real voices with a monotonous synthetic voice. I appreciate the auto translated captions sometimes as they let you get the gist of what's going on, but auto dubbing is pretty much never desirable

▲

cman1444 3 hours ago | parent | prev | next [-]

They should just do it on the subset with the most views.

▲

setopt 3 hours ago | parent [-]

That makes sense actually. Re-sub videos when they cross a million views for example.

▲

cman1444 3 hours ago | parent [-]

Or, cynically, for creators who pay an extra fee.

	▲	computably 3 hours ago \| parent [-]
		YT's business model is monetizing creators' output. Producing a video that gets a million views and generates a bunch of ad revenue is the fee.

▲

kmfrk an hour ago | parent | prev | next [-]

I also suspect that the quality of transcription depends on the popularity of the video. So maybe it's decent for videos with 000,000s of views. Same way the video codec is changed from AVC to VP9 - and occasionally AV1 nowadays.

▲

andai 3 hours ago | parent | prev [-]

I thought so, too. But they also let you create AI dubs of your videos into an arbitrary number of languages?

	▲	jgalt212 3 hours ago \| parent [-]
		opt-in vs opt-out

▲

hengheng 2 hours ago | parent | prev | next [-]

Pixel Phone's locally generated real-time subtitles are better than the ones on YouTube, that's how much effort is put into that.

▲

debugnik 3 hours ago | parent | prev | next [-]

I suspect they're using some new cheap multi-modal model. A couple of times now I've seen the generated subtitles fill muffled speech with text about something unrelated happening on screen.

▲

GuB-42 3 hours ago | parent | prev [-]

YouTube subtitles usually are not that bad, but they tend to get wrong on key moments.

For example the sentence may be "They decided to rewrite the application in Rust", and the subtitle be "They decided to rewrite the application in rats". They made a single mistake, and it is on the most important word. I guess it can be explained by the fact that these key words tend to be uncommon while the rest of the sentence just flows naturally, that is, you barely need to listen, you can guess most of it.

It worries me a bit because as models get better, they may come up with more and more consistent, but wrong transcriptions. Like a typical LLM hallucination, so confidently wrong that you don't question it.

	▲	ComputerGuru 2 hours ago \| parent \| next [-]
		So you’re saying.. they are that bad, actually?
	▲	mrkn1 40 minutes ago \| parent \| prev [-]
		you can transcript fast for free on cpu with yapsnap (github repo)