I really want to know how he solved this problem, which I also face:
>last year i bought a Rodecaster Duo to solve some audio woes to allow myself and my girlfriend to have microphones to our respective computers when gaming together and talking on discord in the same room without any echo
the rodecaster can connect to two computers, and we are both generally in the same discord call. so we have both microphones routed into one input for a computer, and the other person joins with their mic muted and the audio just comes from one client. since the mixing is local there's no echo. email me if you have more questions :)
So both mics will pick up both people (at least somewhat, in the same room) - but because there is no, I assume 20-100ms latency going through the system, to discord, and back - it avoids a slight difference in timing of the two mics picking up the same sound slightly differently. Is that right?
also the audio output of each computer is routed thru the box as well, so i can mix my girlfriend’s computer into her headphones as well as my microphone, so she can hear me with noise canceling headphones, or turn off my microphone if i’m working so she can do stuff without my mic in her ears.
Or if she’s watching a movie or something I can also add her computer audio to my headphones. There’s even a separate audio output for host 1 where you can put ‘chat’ on, like discord on a dedicated interface, so that your application audio is clear and isolated. It’s hella expensive but it really is a great device
It saves on rewiring stuff. Maybe there's only one person talking today. Maybe they're using PC A, or perhaps they're using PC B instead.
Or maybe there's two people in the room, each on different channels altogether. In this case the other person is just uncorrelated background noise instead of a persistent echo.
Or, in-context: There's two people in the same room, both talking on the same Discord channel.
Anyway, audio routing is useful. Being able to route audio with two different PCs is a pretty neat feature of the rodecaster.
Not in the same league or form factor, but I have an old Jabra 65 headset, and the noise canceling is amazing. I can be playing my cello while unmuted on a call, and nobody can hear it.
I know headsets aren't everyone's cup of tea, but a mic close to the source (your mouth) with good noise canceling is a solid solution.
When I read this, I thought that the author would be calculating the derivative of the potential energy with respect to nuclear charge for some reason. Here, the derivative being evaluated appears to be the atomic forces, using jax magic, as opposed to the standard Hellmann-Feynman theorem approach.
Calculating the derivative of the energy with respect to nuclear charge would be fun, as it would let you perform some type of "alchemy" smoothly changing from one element to another. I'm not sure that has any practical use.
Since the gradients are computed with jax, the library can be used to differentiate with respect to all inputs including nuclear charge and basis set parameters (exponents and contraction coefficients). I agree that computing gradients for the nuclear charges could be interesting in the context of molecular design.
But for the colab demo I thought that sticking to nuclear positions (i.e atomic forces) would be easier to visualize.
I think that the audio we are listening too is from some ground recording station that isn't necessarily near the airport. We aren't listening to a recording of what the pilots heard or what air traffic control hears.
Yup. Audio is sourced from volunteers with home radios & antennas. Quality will be dependant on how far the nearest one was from LAX and their personal setup. Not necessarily representative of quality that the controller/pilot was hearing.
Technically correct...and yet let's not kid ourselves that even with a nice radio and a pair of Bose A30's, it can be hard to tease out exactly what's being said when in a congested airspace.
I can't believe how affordable Hetzner is. I just rented a bare metal 48 core AMD EPYC 9454P with 256 GB of ram and two 2 TB NVME ssds for $200/month (or $0.37 per hour). Its hard to directly compare with AWS, but I think its about 10x cheaper.
I play fortnite and marvel rivals with my family. We have lots of fun. I think this genre of game is fantastic if you play with people you know on voice comms. "Solo queuing" in these types of games is not fun for me at all, so I get what you are saying, but they are popular for a reason!
Do you think that you can use those machines for confidential workflows for enterprise use? I'm currently struggling to balance running inference workloads on expensive AWS instances where I can trust that data remains private vs using more inexpensive platforms.
Of course you cannot use these machines "for confidential workflows for enterprise use", at least with AWS you know whose computer you're working with, but also keep in mind that it's really hard to steal your data as long as your data stays in memory, and you use something like mTLS to actually get it in and out of memory via E2EE. You can figure out the rest of your security model along the way, but anything sensitive (i.e. confidential) would surely fall way out of this model.
Just currently exploring how custom AI workflows (e.g. text to sql, custom report generation using private data) can help given the current SOTA. Looking to develop tooling over the next 3-6 months. I'd like to see what we can come up with before dropping $50-100k on hardware.
I threw together a toy project to see if it would help me understand the basic concepts and my takeaway was that, if you can shape your input into something a dedicated classification model (e.g. YOLO for document layout analysis) can work with, you can farm each class out to the most appropriate model.
It turns out that I can run most of the appropriate models on my ancient laptop if I don't mind waiting for the complicated ones to finish. If I do mind, I can just send that part to OpenAI or similar. If your workflow can scale horizontally like my OCR pipeline crap, every box in your shop with RAM >= 16GB might be useful.
Apologies if this is all stuff you're familiar with.
Just in case anyone else reads this and is confused. MRIs only use radio waves. No ionizing (or visible or even IR) radiation is used. The strong magnetic fields are a risk (due to interacting with metallic items embedded in the body). The contrast agents also can cause some undesirable side effects.
I agree that DFT is an approximate solution to the Schrodinger equation, but what would you like to see them do? Quantum monte carlo or configuration integration? These methods do not scale well especially when heavy elements are involved. DFT is the current sweet spot for accuracy vs computational complexity in this field. Making DFT better has been an on-going effort for the 30-40 years at least. It is not an easy task. For many real world materials, DFT is the best we can do.
I like this paper and it appears to be one of the best in the literature so far for AI for materials. Even DFT is not really scalable for this, computing the ground state of even a dozen unit cells requires many many CPU-hours. They themselves in fact relax the proposed structures by minimizing the energy of psuedopotentials, for even DFT is too expensive for that step. I said already I think improving DFT itself is the most potentially impactful application of AI in this space, in my opinion. Of course approximations are always necessary, I’m not at all against that, but DFT ignores or approximates correlations by design so there is an inherent limitation there, which means, if you train your models to predict that, it will have the same limitation. It’s just like with LLMs, only imagine training principally on synthetic data. Obviously LLMs have succeeded with limitation sources of synthetic data, but they are principally trained with “real” data.
I'm in a similar boat. A UK bank thinks I'm one of their customers (someone with a similar name). The reply address is no-reply@ and I'm not about to call a foreign bank.
I had the same happen with a AU insurance company that also made it hard to reach them.
I sent an email to their regulator that this company keeps sending me confidential information about one of their clients. It took one day until I received an email from the company informing me that they've corrected the mistake and I shall no longer receive any emails, and it worked, I haven't received a single one since.
If I made a mistake while entering data, I'd be happy if someone told me they receive emails from me that they probably shouldn't be getting, so I do the same when it's not obvious spam/scam.
same. although, if it's a reservation you're being sent, you can cancel it to let the person know they're using the wrong email (plausible deniability because you don't recognize it, yet are getting a reservation)
Getting a U.K. bank account without having a U.K. mailing address isn't the
easiest thing in the world to do. Maybe someone would be interested
in acquiring it from you.
I use fusion 360 several times a week, but I'm not quite able to follow what you said. Can you provide an example or a link to where they announced this feature?
>last year i bought a Rodecaster Duo to solve some audio woes to allow myself and my girlfriend to have microphones to our respective computers when gaming together and talking on discord in the same room without any echo
reply