Even in the commercial space, there’s a lack of production grade ASR APIs that support diarization and word level timestamps.
My experiences with Google’s Chirp have been horrendous, with it sometimes skipping sections of speech entirely, hallucinating speech where the audio contains noise, and unreliable word level timestamps. And this all is even with using their new audio prefiltering feature.
AWS works slightly better, but also has trouble with keeping word level timestamps in sync.
Whisper is nice but hallucinates regularly.
OpenAI’s new transcription models are delivering accurate output but do not support word level timestamps…
A lot of this could be worked around by sending the resulting transcripts through a few layers of post processing, but… I just want to pay for an API that is reliable and saves me from doing all that work.
I’ve found CIS Controls v8.1 to be good and sane, with actual benefits to security. Level 1 is a solid base, and Level 2 is good for picking from depending on where risks exist in your business.
CIS Benchmarks are worth a look too: They’re best practices for securing typical cloud platforms, SaaS and OS.
Can you write a bit more about your workflow? I've been thinking about doing the same, but since I'm very non-interior-design minded have struggled to ask the right things.
Like...
What are your inputs to the model? Empty renders of the space, or more fully decorated views/ photos?
Do you have a light harness around this to help you discover the style you like and then stay consistent with it?
Do you find that giving a lot of context around the space you're designing helps (it hasn't in my attempts)?
I started with sketchup to make basic floor plans and house shapes. I had a rough idea of the style of the home. I picked "Transitional English Estate" since the build site is out on a farm that sorta looks like the Cotswolds. I used AI in this process to get rough renders and feedback on the floorplan. I then took that basic floorplan and house dimensions to a Draftsman who did a lot of tweaking to get it up to code and fix issues. I got his plans and took it to a Sketchup Pro on Fivver . They made a detailed sketchup model. I then took that model and took screenshots from different perspectives and tweaked the prompt to get renders I liked. These changes were reencorprated into the blueprints. I did the same thing with the interior. Took screenshots from sketchup and put them into AI and tweaked the prompt. https://imgur.com/a/lSIYTYr
Mine was far more lightweight, but u just uploaded pics of my yard and prompted manually a bunch of times. Sometimes id find reference images to give as context, draw on the image to call out specific areas, etc.
It wouldn’t show me the exact things I wanted, but got close enough that I could test ideas and iterate quickly.
We had intense aurora in Berlin, Germany. Green clouds dancing in the sky levels. Started around 22:10 local time or a bit earlier, and at this point there's only a faint red/green glow remaining.
Just spend an hour outsite (Northern Germany, 01:00 MET). Unfortunately nothing to report, neither visual nor on camera.
Maybe I just went to late and missed the show.
I hope you habe more luck in Canada and the US!
I tought I was seeing aurora borealis here at 4 am local time in the neighboring Grand Duchy of Luxemburg but it was just visual pollution due to lights from a city.
Friends who live in Kreuzberg and Friedrichshain could see it pretty well. I'm a bit further south within Berlin where skies are minimally darker, but between 10pm and 11pm it was so bright that light pollution didn't matter.
Funnily enough, despite having lots of alerts set up it was my mom who texted me from northern Brandenburg as she spotted it after getting an alert from RegenRadar, of all apps...
It's amazing to hear it's visible in such a big city. I don't have a good intuition for all the metrics that describe how strong this storm was/is, but when put like this it hits home.
Nice to hear earth weather apps also work great in space haha. I'll keep that in mind when I set up my own notifications. Hopefully I have time to look into it before the next storm hits.
Anecdotally, my APP3 have worked much better then expected when I worked with a drill and jigsaw this weekend. No weird noises and much more NC than the 2 had.
Might be worth trying different ear tips or asking for a replacement if software updates don’t improve it.
In Germany, regulations around asbestos are quite strict: You're not allowed to "seal it in" by e.g. putting floor leveling compounds on top of old asbestos-containing flooring. If there's asbestos found during renovations, it needs to be removed professionally.
When we re-did our kitchen we found asbestos-containing glue under a new-ish layer of tiles one of the previous owners of the apartment just laid on top. I wish regulations would already have been stricter back then (and that they would have been followed - another story...) as this surprise find caused massive delays to the construction and forced us to temporarily move out during the removal and decontamination.
One of the best ways to make DIY-ing in buildings built before 1994 (when asbestos was banned for construction in Germany) safe has been to buy a H-filtration class shop vac. It can filter out asbestos fibers and many other fine dust particles that aren't healthy to inhale and was barely more expensive than a comparably good vac.
Generally yes. I asked my primary care physician and would have been able to get the vaccine dose from the pharmacy (paying for it myself) and she would have administered it.
I am not (usually) photosensitive, but the animated static noise on your websites causes noticable flickering on various screens I use and made it impossible for me to read your article.
For better accessibility and a safer experience[1] I would recommend not animating the background, or at least making it easily togglable.
Edited to add: I am, in fact, photosensitive (due to a genetic retinal condition), and for my eyes, your site as it is very easy to read, and the visualizations look great.
Please let me know what you would like to see more of. Evals are something we take serious, I think this post was ok enough given our constraints, but I'd like to produce content people find useful and I think we can do a lot better.
These cuts have been in progress for months, and it’s a sad state of the world when scientific research and its products are called unscientific and not evidence based without any substantiation.
I’m reminded of this quote from Carl Sagan’s Demon-Haunted World: Science as a Candle in the Dark, as we creep closer and closer to future he describes.
> We've arranged a global civilization in which most crucial elements profoundly depend on science and technology. We have also arranged things so that almost no one understands science and technology. This is a prescription for disaster. We might get away with it for a while, but sooner or later this combustible mixture of ignorance and power is going to blow up in our faces.
Some of those in power do understand science and technology, and have responsible stances.
Some … understand …, and have intentionally-irresponsible stances.
Some do NOT understand, and choose positions either based upon bad information or other priorities entirely.
And those of us not in power are routinely lied to by all four groups, making us question the reputation of literally everything.
I think the end state of this is sort of the dragon eating its tail—not only do those in power no longer understand science and technology (or use their understanding to manipulate others), but the disease then spreads to most everyone else.
Slight addition: some understand that science is immutable (in the platonic sense, whether or not current knowledge is accurate) and therefore willingly work to destroy public belief in science because it's a threat to their own power to declare truth.
Closer to home, here on HN (and in SV generally) there is a lot of support for this regime.
It's infuriating that people who really should know better either remain willfully ignorant or simply view the loss to humanity as an acceptable price to pay for ending up with a lower marginal tax bracket.
Souter warned of the lack of civic education as well. We are just ignorant all around these days!
“An ignorant people can never remain a free people. Democracy cannot survive too much ignorance.”
“I don’t believe there is any problem of American politics in American public life which is more significant today than the pervasive civic ignorance of the Constitution of the United States and the structure of government.”
Theres such a huge focus on STEM yet i rarely see the scientific method or critical thinking and logic being taught. Its like forming your own ideas is no longer a thing and so long as you have the solution you wont need anything original
I see it growing worse in my job where newer staff rarely can come up with new ideas and the older staff are having to hand hold them. They have trouble even stating the problem at times just “i dont know how to fix any of it”. Eventually the solution is either crazy convoluted (a factory for a class for a static function that returns a string of static json) or just crazy in general (let me put this json into an env variable so its now its available global)
Partly this because there was never a focus on STEM there is a focus on a limited subset of T and E that businesses find profitable.
Even then, you need history and philosophy of science classes to actually contextualize and contrast "scientific thinking" to actually get people to go engage with science as an ongoing endeavor rather than a series of concepts from a text book that must be memorized and mastered.
I am citing past comments to demonstrate the individual has a pattern of disregarding basic hygiene measures. And that it's related to their current view on vaccines
And as you can see from their reply they don't mind talking about it.
The phrase "disregarding basic hygiene measures" is also a rhetorical interpretation that indicates you're more interested in battle than curious conversation.
And just because somebody doesn't complain about you digging up old comments, it doesn't make it OK to do. We can't know in advance how it will make someone feel, nor how they actually feel about it, even if they haven't complained about it.
The central issue is still that you seem to be using HN primarily for ideological battle, which is not what it is for.
Hacker News doesn't allow old threads to deprecate and doesn't allow you to delete your account. You choose to create a permanent public record of everything everyone has ever said on this forum, and I've seen mods go back months into a commenter's history to do exactly that.
What's good for the goose should be good for the gander.
Anyone can email us to have comments deleted if there are no replies, or anonymized/redacted if they no longer want the comment linked to their identity.
We cite past comments only to demonstrate a pattern when taking action against a user who has a history of guidelines breaches.
It's a long-established norm on HN that weaponizing historical comments in a present-day argument is not considered a fair debating tactic, as we have no idea if the person still agrees with what they posted a long while ago, nor whether the context of the historical comment is relevant to the present topic.
Geese and ganders have nothing to do with it and I don't know why you're using my efforts to uphold fair debating between community members as grounds to attack our moderation.
The energy of "highlight[ing] bad faith actors" is not what we're going for here. The guidelines ask us to "assume good faith", and there are valid reasons why someone's present-day position could (appear to) contradict a historical position. The best reason is that they have undergone some personal growth, which we would hope of most people over a year or more, and we definitely want to allow space for that on HN.
If they do have such evidence it would be worth while to justify their cuts with that evidence. Just because it was rushed out doesn’t mean it wasn’t used by billions of people tens of billions of times with an extraordinarily low incidence of side effects observed in situ. This is a situation of unparalleled prevalence, including probably everyone reading this now. If there were evidence of problem it would be globally observable and impossible to suppress. Even if the pharma industry would want to, all it would take is a single professor of medicine, statistician, epidemiologist, etc anywhere on earth looking for tenure to be able to produce a compelling and well structured survey and / or meta analysis that showed mRNA vaccines had any level of harm. But it didn’t happen. Because they are safe, and literally everyone who has had one - which is pretty much literally everyone - knows this. RFK Jr knows this too, but he is seeking something other than truth here.
No, what I said about masking policy during COVID was this:
> Reasonable at first. Once it was known that the virus is transmitted via aerosols then forcing masks was the wrong thing to do, specially on children. Useless and damaging.
I don't know why you wouldn't wear a mask to stop aerosols. Masks were a correct answer. The problem was pushing those pieces of crap made out of cut up t-shirts and other hopium, to help cope with the presidential dereliction of duty failing to address the shortage of real PPE P95/P100s.
> I don't know why you wouldn't wear a mask to stop aerosols.
Because the masks don't stop aerosols. They're too tiny for them. Then there's also the fact that being so light they accumulate in the air and last there for an hour or longer and can even infect people through the eyes.
That argument was always trotted out in the context of the hopium made out of t-shirts, or surgical masks with open flow paths around the filter. Everything I saw confirmed that P100's were quite effective at the relevant drop size (as you'd expect them to be, as they stop things like atomized and aerosolized paint spray). As far as the eyes, reducing viral load is worthy on its own. Or if you want the proper solution, that is a full face mask. But either way, not having a perfect solution certainly doesn't indicate going in the opposite direction and completely giving up on the idea.
> not having a perfect solution certainly doesn't indicate going in the opposite direction and completely giving up on the idea.
Having a so-so solution can in fact be worse than nothing since it makes people think they're safe, which leads to much greater exposure than if they thought they were in danger. Promoting ventilation and open-air interaction is a much more effective solutions but the latter was banned and even persecuted.
You didn't address my core point, and you're still pulling in this odd direction of pointing to the failings of the straightforward approach to justify contrarianism based around doing even less.
Ventilation and open-air interaction were also both widely promoted, and widely used for things like social gatherings in parks and outside of homes. There were plenty of situations (eg grocery stores) where they were inapplicable from an individual perspective, hence the masks.
I'm glad this is being done. Government funded research like this is just another financial engineering opportunity for these research companies/universities. I am not making this up, just check the other post about scientific fraud at scale which is currently on the front page of HN https://news.ycombinator.com/item?id=44796526
All we're doing is shifting your tax dollars to the military, which is another financial engineering opportunities for defense contractors.
If my money's going to go somewhere I'd rather it go to research universities/companies and get outputs that will save lives rather than destroy lives.
Privatizing vaccine development doesn't remove the incentives for scientific fraud. If anything, we trust the government with these incalculably valuable public-works projects because we cannot afford to write off the alternative. Not for COVID, not for the seasonal flu, and not for the next plague/ebola/malaria of your lifetime either.
> Privatizing vaccine development doesn't remove the incentives for scientific fraud.
Right it won't. My point however is that don't create an avenue, private or public, which has dedicated funds for such activities. It just ends up becoming an industry which tries to gain off those funds than doing any real research.
That's how all industry works. The side that invests most in innovative techniques is rewarded with larger margins and incentivized to wring out every last dollar they can.
The "if we don't look at the research then it doesn't exist" mentality is why TSMC is 5 generations ahead of Intel's fabs.
My experiences with Google’s Chirp have been horrendous, with it sometimes skipping sections of speech entirely, hallucinating speech where the audio contains noise, and unreliable word level timestamps. And this all is even with using their new audio prefiltering feature.
AWS works slightly better, but also has trouble with keeping word level timestamps in sync.
Whisper is nice but hallucinates regularly.
OpenAI’s new transcription models are delivering accurate output but do not support word level timestamps…
A lot of this could be worked around by sending the resulting transcripts through a few layers of post processing, but… I just want to pay for an API that is reliable and saves me from doing all that work.
reply