Hacker Newsnew | past | comments | ask | show | jobs | submit | listic's commentslogin

build-your-own-x*, a popular compilation of "well-written, step-by-step guides for re-creating our favorite technologies from scratch", contains some.

* https://github.com/codecrafters-io/build-your-own-x


How easy is it to become their (Cerebras) paying customer? Last time I looked, they seemed to be in closed beta or something.

I signed up and got access within a few days. They even gave me free credits for a while

That's gone now. They do drops from time to time, but their compute platform is saturated.

Apparently, yes. Apparently, they present a curated set of texts.


Thanks for the advice! Could you please share how did you enable voice transcription for your setup and what it actually is?


I use https://github.com/braden-w/whispering with an OpenAI api key.

I use a keyboard shortcut to start and stop recording and it will put the transcription into the clipboard so I can paste into any app.

It's a huge productivity boost - OP is correct about not overthinking trying to be that coherent - the models are very good at knowing what you mean (Opus 4.5 with Claude Code in my case)


I just installed this app and it is very nice. The UX is very clean and whatever I say it transcribes it correctly. In fact I'm transcribing this comment with this app just now.

I am using Whisper Medium. The only problem I see is that at the end of the message it sometimes puts a bye or a thank you which is kind of annoying.


I am all ready to believe that with LLMs it's not worth it trying to be too coherent: I did successfully use LLMs to make sense of what incoherent-sounding people say. (in text)


Aquavoice, YC company, really good. Got it after doing a bit of research on here, there's something for Mac that's supposed to be good too.

If you want local transcription, locally running models aren't quite good enough yet.

They use right-ctrl as their trigger. I've set mine to double tap and then I can talk with long pauses/thinking and it just keeps listening till I tap to finish.


I'm using Wispr flow, but I've also tried Superwhisper. Both are fine. I have a convenient hotkey to start/end recording with one hand. Having it just need one hand is nice. I'm using this with the Claude Code vscode extension in Cursor. If you go down this route, the Claude Code instance should be moved into a separate window outside your main editor or else it'll flicker a lot


another option is MacWhisper if someone is on macOS and doesn't want to pay for subscription (just one time payment) - pretty much all of those apps these days use paraspeech from NVIDIA which is the fastest and the best open source model that can run on edge devices.

Also haven't tried but on latest MacOS 26 apple updated their STT models so their build in voice dictation maybe is good enough.


For me, on Mac, VoiceInk has been top notch. Got tired of superwhispr


Spokenly on macOS with Soniox model.


For the one disinclined to get into closed source, proprietary tools, what is the next best thing to try?

I heard of Cline and Aider, but didn't try anything.


I doubt there will ever be a market for a device with this specific (or many other kinds of other specific) set of requirements, especially if one insists on that 'last bit of polish'. Short of fully developed molecular nanotechnology or similar, allowing economic manufacturing of bespoke hardware in single copies, which is another world from the one we live in, I don't see such wishes come true, so the author (and others like him) will have to settle for something else.


> If you think you (or a loved one) may have a psychological condition, go to a psychologist and get a screening. The diagnosis isn't the important part. The value is in the 20-something pages of detailed analysis by a professional.

Throughout my entire interaction with psychiatry (years, on and off) I never figured this is a thing. Go figure.


Funny thing, taking into account the state of affairs in the country: CachyOS mirror for Russia is https://archlinux.gay

(I am Russian; left the country in 2022)


Sometimes, they return.

Recently, I read a post by an online musician friend that someone made a tip of 5 rubles with a banknote in Armenia - too low a sum to bother exchanging it in a foreign country. I was perplexed [1] because I don't recall ever [2] seeing such banknote - the lowest I remember seeing was 10 - and found that yes, it was introduced in 1998, discontinued and withdrawn from circulation in 2001 due to inflation and apparently reintroduced (maybe briefly?) in December 2022/January 2023.

I don't know why.

[1] https://www.perplexity.ai/search/in-post-soviet-russia-when-...

[2] I am Russian; born in 1977 and left the country in December, 2022.


Not to defend what Altman is saying, but is OpenAI actually using or going to use that much power? This Reuters source the US power budget will be 4,187 TWh in 2025: https://www.reuters.com/business/energy/us-power-use-reach-r...


The US is the second most energy-hungry country on Earth (the first one being China, a country that accounts for almost 30% of the global manufacturing output), so I think it's better to compare it against the rest of the world:

https://en.m.wikipedia.org/wiki/List_of_countries_by_electri...

Every country bellow the top 40 has a consumption lower than 87.6 TWh per year, that includes developed countries like Finland and Belgium, so yes, 10 gigawatts is a lot of power.


Adding 10GW of power as they plan, is 87TWh. 2%.

That’s huge. The hope is that this can drive renewables or nuclear rollout, but not sure that hope is realistic.


Altman is an investor in OKLO. I'm expecting an announcement with companies like that where they can federal dollars to speed up R&D.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: