A small lathe built in a Japanese prison camp (1949)
525 by CommieBobDole | 144 comments on Hacker News.
Tuesday 30 April 2024
Monday 29 April 2024
Sunday 28 April 2024
Saturday 27 April 2024
Friday 26 April 2024
Thursday 25 April 2024
Wednesday 24 April 2024
Tuesday 23 April 2024
Monday 22 April 2024
Sunday 21 April 2024
Friday 19 April 2024
Thursday 18 April 2024
Wednesday 17 April 2024
Tuesday 16 April 2024
Monday 15 April 2024
Sunday 14 April 2024
Friday 12 April 2024
New best story on Hacker News: Show HN: Sonauto – A more controllable AI music creator
Show HN: Sonauto – A more controllable AI music creator
441 by zaptrem | 230 comments on Hacker News.
Hey HN, My cofounder and I trained an AI music generation model and after a month of testing we're launching 1.0 today. Ours is interesting because it's a latent diffusion model instead of a language model, which makes it more controllable: https://sonauto.ai/ Others do music generation by training a Vector Quantized Variational Autoencoder like Descript Audio Codec ( https://ift.tt/DcbQ5XO ) to turn music into tokens, then training an LLM on those tokens. Instead, we ripped the tokenization part off and replaced it with a normal variational autoencoder bottleneck (along with some other important changes to enable insane compression ratios). This gave us a nice, normally distributed latent space on which to train a diffusion transformer (like Sora). Our diffusion model is also particularly interesting because it is the first audio diffusion model to generate coherent lyrics! We like diffusion models for music generation because they have some interesting properties that make controlling them easier (so you can make your own music instead of just taking what the machine gives you). For example, we have a rhythm control mode where you can upload your own percussion line or set a BPM. Very soon you'll also be able to generate proper variations of an uploaded or previously generated song (e.g., you could even sing into Voice Memos for a minute and upload that!). @Musicians of HN, try uploading your songs and using Rhythm Control/let us know what you think! Our goal is to enable more of you, not replace you. For example, we turned this drum line ( https://ift.tt/6n90I31 ) into this full song ( https://ift.tt/5wPvMUk skip to 1:05 if impatient) or this other song I like better ( https://ift.tt/faT9s1u - we accidentally compressed it with AAC instead of Opus which hurt quality, though) We also like diffusion models because while they're expensive to train, they're cheap to serve. We built our own efficient inference infrastructure instead of using those expensive inference as a service startups that are all the rage. That's why we're making generations on our site free and unlimited for as long as possible. We'd love to answer your questions. Let us know what you think of our first model! https://sonauto.ai/
441 by zaptrem | 230 comments on Hacker News.
Hey HN, My cofounder and I trained an AI music generation model and after a month of testing we're launching 1.0 today. Ours is interesting because it's a latent diffusion model instead of a language model, which makes it more controllable: https://sonauto.ai/ Others do music generation by training a Vector Quantized Variational Autoencoder like Descript Audio Codec ( https://ift.tt/DcbQ5XO ) to turn music into tokens, then training an LLM on those tokens. Instead, we ripped the tokenization part off and replaced it with a normal variational autoencoder bottleneck (along with some other important changes to enable insane compression ratios). This gave us a nice, normally distributed latent space on which to train a diffusion transformer (like Sora). Our diffusion model is also particularly interesting because it is the first audio diffusion model to generate coherent lyrics! We like diffusion models for music generation because they have some interesting properties that make controlling them easier (so you can make your own music instead of just taking what the machine gives you). For example, we have a rhythm control mode where you can upload your own percussion line or set a BPM. Very soon you'll also be able to generate proper variations of an uploaded or previously generated song (e.g., you could even sing into Voice Memos for a minute and upload that!). @Musicians of HN, try uploading your songs and using Rhythm Control/let us know what you think! Our goal is to enable more of you, not replace you. For example, we turned this drum line ( https://ift.tt/6n90I31 ) into this full song ( https://ift.tt/5wPvMUk skip to 1:05 if impatient) or this other song I like better ( https://ift.tt/faT9s1u - we accidentally compressed it with AAC instead of Opus which hurt quality, though) We also like diffusion models because while they're expensive to train, they're cheap to serve. We built our own efficient inference infrastructure instead of using those expensive inference as a service startups that are all the rage. That's why we're making generations on our site free and unlimited for as long as possible. We'd love to answer your questions. Let us know what you think of our first model! https://sonauto.ai/
Thursday 11 April 2024
New best story on Hacker News: Show HN: I made a new sensor out of 3D printer filament for my PhD
Show HN: I made a new sensor out of 3D printer filament for my PhD
530 by 00702 | 99 comments on Hacker News.
Here's a "behind-the-scenes" look at my development of a cool sensor during my PhD (electrical engineering). This sensor is only about 1/3 of my total research for my degree and took about a year. I've been on HN for a while now and I've seen my fair share of posts about the woes of pursuing a PhD. Now that I'm done with mine I wanna share some anecdotal evidence that doing a PhD can actually be enjoyable (not necessarily easy) and also be doable in 3 years. When I started I knew I didn't want to work on something that would never leave the lab or languish in a dissertation PDF no one will ever read. Thanks to an awesome advisor I think I managed to thread the needle between simplicity and functionality. Looking back, the ideas and methods behind it are pretty straightforward, but getting there took some doing. It’s funny how things seem obvious once you've figured them out! Oh, I love creating GUIs for sensor data and visualizations as you'll see -- it's such a game changer! pyqtgraph is my go-to at the moment - such a great library.
530 by 00702 | 99 comments on Hacker News.
Here's a "behind-the-scenes" look at my development of a cool sensor during my PhD (electrical engineering). This sensor is only about 1/3 of my total research for my degree and took about a year. I've been on HN for a while now and I've seen my fair share of posts about the woes of pursuing a PhD. Now that I'm done with mine I wanna share some anecdotal evidence that doing a PhD can actually be enjoyable (not necessarily easy) and also be doable in 3 years. When I started I knew I didn't want to work on something that would never leave the lab or languish in a dissertation PDF no one will ever read. Thanks to an awesome advisor I think I managed to thread the needle between simplicity and functionality. Looking back, the ideas and methods behind it are pretty straightforward, but getting there took some doing. It’s funny how things seem obvious once you've figured them out! Oh, I love creating GUIs for sensor data and visualizations as you'll see -- it's such a game changer! pyqtgraph is my go-to at the moment - such a great library.
Wednesday 10 April 2024
New best story on Hacker News: Show HN: We built PriceLevel to find out what companies pay for SaaS
Show HN: We built PriceLevel to find out what companies pay for SaaS
453 by cluo21 | 126 comments on Hacker News.
Hey HN! Christine and Steven here. As a PM and engineer, we’ve both evaluated and purchased a lot of software. One of the biggest frustrations was figuring out how much it would cost us without having to go through the sales process. When we did have a quote, we had no idea if we were getting a good deal or ripped off. We built a site where you can see what other companies are actually paying for SaaS and enterprise software. Buyers contribute prices via quotes, pricing proposals, and other documentation to ensure quality. We unlocked Talkdesk for Show HN users so that you can use the product without needing to sign in or upgrade. Check it out at https://ift.tt/vXa78dZ . Would love to hear any feedback, thank you!
453 by cluo21 | 126 comments on Hacker News.
Hey HN! Christine and Steven here. As a PM and engineer, we’ve both evaluated and purchased a lot of software. One of the biggest frustrations was figuring out how much it would cost us without having to go through the sales process. When we did have a quote, we had no idea if we were getting a good deal or ripped off. We built a site where you can see what other companies are actually paying for SaaS and enterprise software. Buyers contribute prices via quotes, pricing proposals, and other documentation to ensure quality. We unlocked Talkdesk for Show HN users so that you can use the product without needing to sign in or upgrade. Check it out at https://ift.tt/vXa78dZ . Would love to hear any feedback, thank you!
Tuesday 9 April 2024
Monday 8 April 2024
Saturday 6 April 2024
New best story on Hacker News: Ask HN: What is the most useless project you have worked on?
Ask HN: What is the most useless project you have worked on?
584 by panqueca | 638 comments on Hacker News.
If you're feeling useless, remember that I exist. Let me give you some context. I work in the pipeline automation department of a company. Last month, our team decided to deprecate an internal tool due to several maintenance issues. So we created a pipeline that automates the implementation of this legacy tool, in case other teams needed to use it. (WHAT???) This month, a guy in my team found some improvement scenarios in the automation. So I was chosen to implement this changes in this legacy internal tool. The thing is, after I finished the adjustments, my pull requests are not getting approved due to adjustments meticulously requested by this guy in my team. Adjustments to make the pipeline automation even more resilient in complete unlikely scenarios. But this same week, my TL sent notices to all the other teams informing them that this internal tool has been deprecated and they should no longer use it. So what sense does it make to have a pipeline automation that implements the use of the deprecated tool? And if it has been deprecated, why would I need to make an adjustment for the automation to be even resilient if no one should be able to use it anymore? So why am I being allocated to work on in such waste of time like it? (WTF???) This makes me wonder, how many people have to work on something that they see no sense in doing at all. So once again, if you're feeling useless, remember that I exist.
584 by panqueca | 638 comments on Hacker News.
If you're feeling useless, remember that I exist. Let me give you some context. I work in the pipeline automation department of a company. Last month, our team decided to deprecate an internal tool due to several maintenance issues. So we created a pipeline that automates the implementation of this legacy tool, in case other teams needed to use it. (WHAT???) This month, a guy in my team found some improvement scenarios in the automation. So I was chosen to implement this changes in this legacy internal tool. The thing is, after I finished the adjustments, my pull requests are not getting approved due to adjustments meticulously requested by this guy in my team. Adjustments to make the pipeline automation even more resilient in complete unlikely scenarios. But this same week, my TL sent notices to all the other teams informing them that this internal tool has been deprecated and they should no longer use it. So what sense does it make to have a pipeline automation that implements the use of the deprecated tool? And if it has been deprecated, why would I need to make an adjustment for the automation to be even resilient if no one should be able to use it anymore? So why am I being allocated to work on in such waste of time like it? (WTF???) This makes me wonder, how many people have to work on something that they see no sense in doing at all. So once again, if you're feeling useless, remember that I exist.
Thursday 4 April 2024
New best story on Hacker News: Show HN: I've built a locally running Perplexity clone
Show HN: I've built a locally running Perplexity clone
490 by nilsherzig | 99 comments on Hacker News.
The video demo runs a 7b Model on a normal gaming GPU. I think it already works quite well (accounting for the limited hardware power). :)
490 by nilsherzig | 99 comments on Hacker News.
The video demo runs a 7b Model on a normal gaming GPU. I think it already works quite well (accounting for the limited hardware power). :)
Wednesday 3 April 2024
New best story on Hacker News: 'Lavender': The AI machine directing Israel's bombing in Gaza
'Lavender': The AI machine directing Israel's bombing in Gaza
548 by contemporary343 | 482 comments on Hacker News.
548 by contemporary343 | 482 comments on Hacker News.
Tuesday 2 April 2024
Monday 1 April 2024
Subscribe to:
Posts (Atom)
-
Path to a free, self-taught education in Computer Science 495 by netfortius | 130 comments on Hacker News.
-
U.S. Postal Service starts nationwide electric vehicle fleet, buying 9,250 EVs 444 by lxm | 336 comments on Hacker News.
-
So you want to study mathematics 507 by musgravepeter | 292 comments on Hacker News.