TachyonicBytes

no profile record

comments

tahun laludiscuss

It has the payment button at least

tahun laludiscuss

It seems that both use / leverage pyannote. I wonder if the whisperX pipeline can be combined with DiCoW-v2.

tahun laludiscuss

Yes, this is exactly where I am going. The LLM also has an advantage, because you can give it the context of the audio (e.g. "this is an audio transcript from a radio show about etc. etc."). I can foresee this working for a future whisper-like model as well.

There are two ways to parse your first sentence. Are you saying that you used whisperX and it doesn't do well with diarization? Because I am curious of alternative ways of doing that.

tahun laludiscuss

It seems to be pre-built on github, in releases.

tahun laludiscuss

I use whisperfile[1] directly. The whisper-large-v3 model seems good with non-English transcription, which is my main use-case.

I am also eyeing whisperX[2], because I want to play some more with speaker diarization.

Your use-case seems to be batch transcription, so I'd suggest you go ahead and just use whisperfile, it should work well on an M4 mini, and it also has an HTTP API if you just start it without arguments.

If you want more interactivity, I have been using Vibe[3] as an open-source replacement of SuperWhisper[4], but VoiceInk from a sibling comment seems better.

Aside: It seems that so many of the mentioned projects use whisper at the core, that it would be interesting to explicitly mark the projects that don't use whisper, so we can have a real fundamental comparison.

[1] https://huggingface.co/Mozilla/whisperfile

[2] https://github.com/m-bain/whisperX

[3] https://github.com/thewh1teagle/vibe/

[4] https://superwhisper.com/

tahun laludiscuss

There was an article around here about how battery life actually improves if you ultra-fast charge the battery when you make it.

Maybe it will deteriorate it, but it seems that the effect that different charge types have on batteries may not be complete yet.

tahun laludiscuss

I have to add https://nullprogram.com, just because of the care the author took to have it work better in lynx[1]:

    Just in case you haven’t tried it, the blog also works really well with terminal-based browsers, such as Lynx and ELinks. Go ahead and give it a shot. The header that normally appears at the top of the page is actually at the bottom of the HTML document structure. It’s out of the way for browsers that ignore CSS.

[1] https://nullprogram.com/blog/2017/09/01/

tahun laludiscuss

Is this a different method from the httptap [1] that was on hackernews a few weeks ago? Somebody in that post seemed to say that it also generates CA certificates on the fly.

[1] https://news.ycombinator.com/item?id=42919909

tahun laludiscuss

I use Zotero[1] as a personal web archiver. It downloads the page locally, placing most of the resources inside a single html file (pictures become base64 encoded pngs, for example). I find it the best way to have the content available offline and also to be able to reference it easily, seeing as it is a citation manager first.

[1] https://www.zotero.org/

tahun laludiscuss

It can, but that's another type of "shallow", or more exactly "not-deep" cloning, called blobless cloning [1]. There is also treeless cloning, with other tradeoffs, but much to the same effect.

I found this[2] very enlightening.

[1] https://github.blog/open-source/git/get-up-to-speed-with-par...

[2] https://www.howtogeek.com/devops/how-to-use-git-shallow-clon...

tahun laludiscuss

I assume that framework is not open-source or somewhere I can look at it?

tahun laludiscuss

Not the OP, but they are "Headers". Probably coming from the <h1> tag in html. What outsiders probably call "Headlines".

tahun laludiscuss

What libraries have you seen that do this?

2 tahun yang laludiscuss

You don't need error correction for some crypto primitives. There are QKD networks deployed that don't have that kind of error correction, as far as I know.

2 tahun yang laludiscuss

With quantum tokens, law enforcement have to crack your physical devices, so they at least have to good-old-fashion bug your devices. With classical schemes, they can intercept on the way.

I wouldn't say that current side-channels, most certainly enabled by hardware, not software, are easier to audit.

2 tahun yang laludiscuss

Yes, in theory. In practice, photon generators won't behave perfectly. There are lots of possible attacks, like photon splitting [1].

[1] https://onlinelibrary.wiley.com/doi/full/10.1002/qute.202300...

2 tahun yang laludiscuss

Would you mind giving us the titles of those books?

2 tahun yang laludiscuss

You can always try infinitesimal analysis[1]

[1] https://people.math.wisc.edu/~hkeisler/calc.html

2 tahun yang laludiscuss

Simulating a quantum system is a hard challenge and it's actually how Feynman proposed the quantum computing paradigm in the first place. It's basically the original motive.

2 tahun yang laludiscuss

Link to the actual article: https://www.nature.com/articles/s41586-024-08449-y