woodson

6 ms·

1.
by woodson 28d ago
This is great, as is the CasioGShockSmartSync Android app. I wish there was an iOS app, as the official app is horrible, only works after signing up for an account, and only works if given every tracking permission possible to vacuum the ma
2.
by woodson 28d ago
To be fair, the hobbyist hardware mod variants that do step tracking only last for days, not years AFAIK.
3.
by woodson 28d ago
Yes, it seems to be a repackaging of the ABL-100WE and has a long battery life. Unfortunately, their app is horrible, only works after signing up for an account, and only works if given every tracking permission possible to vacuum the maxim
4.
by woodson 1mo ago
If all you're looking for is a step counter in your watch, check out Casio ABL100WE. The battery lasts for two years (CR2016).
5.
by woodson 2mo ago
They did. Check out the Phi models (e.g., https://huggingface.co/microsoft/phi-4 ).
6.
by woodson 2mo ago
Try option-click on the green icon in the app window. It’s well hidden and unintuitive, but I think it does what you want.
7.
by woodson 2mo ago
For a long time, AI was a bad word that stood for unfulfilled promise. See AI Winter. Hence, researchers strictly avoided the term while still working on learning algorithms, the same that power LLM training.
8.
by woodson 2mo ago
There’s a bit of a grey area, for example speech recognition. Would you classify that as analytic or generative? Whisper and speech LLMs work pretty well, but can completely make up stuff that wasn’t in the audio at all (see e.g. “thank you
9.
by woodson 2mo ago
Thanks for sharing, this is a great improvement (also and especially for human use! Renders much better on mobile phones compared to Confluence).
10.
by woodson 2mo ago
> Alongside the poorly lit spaceship. Spaceships are workplaces and workplaces should provide adequate illumination so you can see what you are doing. I always thought the same about CSI. Have the set designers ever been to a lab?
11.
by woodson 3mo ago
> 5090 ($2k MSRP but realistically $3-3.5k) These days, more like >$4.1K (at least in the US).
12.
by woodson 4mo ago
Yeah, there have been several papers with attacks on Whisper: - Inject adversarial noise to make it transcribe what you want ( https://arxiv.org/abs/2210.17316 ) - Stop it from transcribing ( https://arxiv.org&
13.
by woodson 4mo ago
Indeed. And the one in Admont, Austria: https://stiftadmont.at/en/about-the-abbey-library/
14.
by woodson 5mo ago
HF is notorious for making it difficult to work offline (or at least not waste time trying to connect when everything needed is offline) and is constantly changing how it is being handled. Previously, there was TRANSFORMERS_OFFLINE, HF_DAT
15.
by woodson 5mo ago
You can use devcontainers (in VSCode or separate), like this: https://github.com/entn-at/claude-rust-devcontainer/ This will limit the agent in what it can do in the system and what IPs/domains it can reach.
16.
by woodson 5mo ago
> But I don't see a scenario where pytorch needs network access. Training models across multiple compute nodes? That’s a big one.
17.
by woodson 5mo ago
Not if one is running it in a non-privileged vm/container with restricted network access. But everything is YOLO these days.
18.
by woodson 5mo ago
Why generative? Or has it been decided that only generative models are “AI”?
19.
by woodson 5mo ago
It highly depends on the sort of data you’re processing (phone calls, podcasts, meetings of more people recorded using single channel?). For NVIDIA/NeMo, check out their softformer diarization models (also streaming).
20.
by woodson 5mo ago
Well, but their customers are those that buy Apple hardware.
21.
by woodson 5mo ago
Look into RWKV.
22.
by woodson 6mo ago
It’s a distinction that IMHO likely doesn’t make much difference, at least for the mostly automated/non-interactive coding agent use case. What matters more is how well the post-training on synthetic harness traces works.
23.
by woodson 6mo ago
There are still opportunities, but they aren’t paid nearly as well as less researchy positions in industry. US post-doc salaries at state universities aren’t that high.
24.
by woodson 7mo ago
You mean Moshi ( https://github.com/kyutai-labs/moshi )? Since Personaplex is just a finetuned Moshi model.
25.
by woodson 8mo ago
…who lives in Oregon, in the US.
26.
by woodson 8mo ago
This is about speech to text, not speech recognition.
27.
by woodson 8mo ago
You still need a tax accountant in France to register the FTE and file paperwork with the tax office and social insurance.
28.
by woodson 8mo ago
The bank issues/refusals may have something to do with FATCA. If you have anything to do with the US in terms of taxes, many EU banks don’t want you as their customer. If it’s a subsidiary of a foreign company, then a lot of paperwork
29.
by woodson 8mo ago
This is basically a devcontainer, right?
30.
by woodson 8mo ago
Jeez, OP asked what it means in this context (zero-shot voice cloning), where you quoted a generic definition copied from Wikipedia. I defined it concretely for this context. Don't take it as a slight, there is no need to get all argum

More ›