Hacker News new | past | comments | ask | show | jobs | submit | hnenjoyer_93's comments login

According to the preview, some of these (30 thousand results?) are "Wikipedia article audios"[1], so text turned into speech and now turned back to text

[1]: https://huggingface.co/datasets/PleIAs/YouTube-Commons/viewe...


emad_9608 on Stable Audio 2.0 release post[1]:

> Team is working on an open version of this for https://github.com/Stability-AI/stable-audio-tools Dataset just taking some time. Lots of improvements to come like speech, customisation, comfy & more.

[1]: https://old.reddit.com/r/StableDiffusion/comments/1buruzc/co...

Edit: reformatted to one paragraph because HN didn't like the newlines


Sorry I totally missed that. Thank you!



I briefly skimmed through the text, and it seems to cover only the anti-AI arguments and how some people agree and disagree. At the same time there seems to be no pro-AI arguments, like improving the efficiency of work, medical advancements and other useful things. Maybe that's what was meant with such a title but I would like the author to cover the whole picture next time


A new winter is never coming, the last few years have shown the potential of machine learning and its use is only going to increase


Machine learning has taken huge leaps but the expectations for it are getting blown out of proportion. The use is going to increase for sure but there are many weaknesses in the technology that tend to be overlooked.

E.g. autonomous driving already proved too hard a task for ML for the foreseeable future. Also "hallucination" is a problem with no clear solution in sight.


>autonomous driving already proved too hard a task for ML for the foreseeable future

Has it? Seems like Waymo is still making good progress. They became generally available in SF recently IIRC.


I've seen hints of this in the past month or so: people who were acting like true general AI happened a year ago now just talking about it as "word generators".

Gemini felt like the tipping point where the flaws became obvious, which they then started noticing in the others.


Considering the amount of carbon being pumped out to power all this, you may be right.


Not open source and you require a subscription to run a scraper from the user's own device?

Edit: I see that it's free for now but still


What's wrong with that?


Aside from what the sister comment said, the app being closed source means we can't reliably check what it actually does, how it uses the data and whether it will get us IP bans on the job websites. Paying a subscription for using our own computer's resources and network to scrape the sites is just wrong, I wouldn't complain if it was some web service running in a cloud that just sent notifications though


> Paying a subscription for using our own computer's resources and network

I agree with the rest of your comment, but not sure this part is accurate. People pay for Microsoft Word or Photoshop etc., and then run them entirely on their own computer using their own computer's resources and network.


I should have really worded it better connecting that to the above statements. The problem is that in this particular case a cloud solution would be a better choice for this exact task (unlike for Word or Photoshop). Also the author said they might consider that in another comment


That makes sense in theory, but I'm not sure most people would bother reading the source code. Regarding your IP banning concerns, since the app only scrapes the links you save it's just like you would manually refresh them in Chrome so don't think this would be an issue (no data to back this up tho, other than my own PC which is running the app continuously and has a ton of links saved for testing).

I made it a desktop app to avoid needing to run expensive cloud resources for scraping (which also includes IP proxies to get around bot detection limits). This would have meant a higher cost for the end user, which I absolutely did not want since it's targeting also unemployed people.

A lot of people have had that complaint and I'm considering building a cloud version with a higher price point, but not a priority right now.


We basically don't know who the author is or what are they intentions, no available info anywhere.


Fair enough and you see open sourcing the code as the only way? Or any other options?


think about the user experience:

1) they go to your website, you have links to social websites like linkedin.

2) user goes to your linkedin page and there's no single person associated with it.

conclusion: using "internet best practices" it sounds like a scam to me. so in a forum like HN i would expected you could open source it _or_ tell who you are, what data is being sent and collected, etc.


never thought about it that way. I'm a shy person so don't usually like putting my name out there. Also, personally, I've never cared who's behind a product, if I like it, I'll use it.


Your personal traits do not invalidate the arguments given to you. You have to admit that not putting any identify forward AND not making it open-source is making it hard to discern from scam services whose only goal is to farm people's emails and web visiting preferences.

I'd like to use your thing but you must show a bit more goodwill first.


Don't get me wrong, but feel free to not use it if that's a MAJOR issue for you. I don't want to make the source code public because it will be a comercial product. As for my personal info, the username I use here is actually my lastname + firstname if you want to dig me up on linkedin. I do not want to put my name on the app, it's just how I am.


I hope you don't get me wrong as well. I and others are pointing out to you that being shy and undercover (so to speak) involuntarily mixes you in with actual shady folk (who keep information vague for darker purposes).

I've bookmarked your project and will keep a look on it. I don't mind paying for convenience if the product works but for now I'll just monitor and see how it develops.


Z8 and Z9 already record 8K60 12 bit RAW internally, but of course more cameras joining the list is welcome


Interesting that some time ago Red filed a lawsuit against Nikon for using a RAW video compression algorithm that Red claimed infringed its patents, but later dismissed it:

https://dpreview.com/news/5338402238/red-is-suing-nikon-for-...

https://dpreview.com/news/9301564383/nikon-denies-red-s-laws...

https://www.newsshooter.com/2023/04/28/red-patent-lawsuit-ag...


Am I reading this right in that Red has somehow managed to patent the idea of compressing 4K RAW footage on device which holds up the whole industry?


Not quite. The claims, in the only one I've seen regarding RAW compression, are for a specific pre-emphasis curve being applied to the raw data, then the raw data being compressed and only cover this being done in a video camera.

When looking at a patent check the "Claims" section. An infringing device would have to perform those steps in the order provided for the patent holder to have a claim.

Disclaimer: I am not a lawyer, but I've had a lot of dealing with patents.


the patent is this one: https://patents.google.com/patent/US7830967B1/en

Which looks like it's patenting debayering. However I'm not a patent lawyer so I don't know.

I suspect that Nikon were using something like jpeg2000 to store raw frames, and thats why RED were getting all pissy.


It’s patenting compressing the raw photo site values before debayering, which saves space and in theory allows for better, non-realtime debayering algorithms (more relevant when the patent was filed).


if its not an actual algorithm, but merely the thought of compressing the raw data (im not saying it is/isnt, I dont know), then its obviously totally ridiculous, and frankly, anyone that would even presume to think one should apply for such a patent should just be taken out back and disposed of. And then of course the patent office aswell


Maybe, Redcode is based on JPEG2000

https://youtu.be/IJ_uo-x7Dc0


> Maybe, Redcode is based on JPEG2000

Oh it is JPEG2000, I _think_ it used to be stored in a tar-like wrapper, but I could be misremembering. however It is 100% JPEG2000.


It must be a specific implementation because Black Magic has had their own brand of this for years and don’t seem to have been targeted by RED, though I may be out of the loop on something?


Blackmagic raw is actually debayered in the camera, which would avoid RED’s “compress raw video by color plane” patent.

> A drawback of Blackmagic RAW over other RAW formats is that it does a partial de-bayer within the camera. This means that you are not actually working with fully raw data from the camera sensor. According to the Blackmagic website the “noise management, sensor profiling and new edge reconstruction algorithms” are part of the partial de-mosaic.


They settled out of court, Nikon started paying licensing fees. Was confirmed by Jarred on a podcast last year.

Maybe that’s what’s partially responsible for the acquisition - also Red struggling to get market share from Arri Alexa on bigger budget productions, and facing pretty stiff competition from lower priced cameras by Blackmagic, Sony etc


Makes sense from a Nikon perspective: I have to pay anyway, over period X that amounts to sum Y, put that next the acquisition price and the estimatebof how long you want to use the tech you are paying for and there is your business case.

Plus, you get new business and a new product range you don't have yet. That alone justifies an acquisition.


Right, its like RED was in the middle market, which is a bad place to be in any product/industry. Too expensive for consumers, but still not able to command premium pricing from top end pros?


Vision Pro is Micro OLED, not microLED


https://chat.lmsys.org/

Choose Arena (side-by-side), it has Claude 3 Opus, Sonnet and GPT-4


Consider applying for YC's W25 batch! Applications are open till Nov 12.

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

Search: