akreal

5 ms·

1.
by akreal 2mo ago
This field got reactivated recently, some new models were released (in the order of release, but the last one seems the best at the moment): 1. ZIPA https://github.com/lingjzhu/zipa 2. POWSM https://huggingf
2.
by akreal 5mo ago
AFAIK all ways of running docker on MacOS rely on a amd64 virtual machine with Linux kernel in it.
3.
by akreal 6mo ago
WhisperX is not a model but a software package built around Whisper and some other models, including diarization and alignment ones. Something similar will be built around the Cohere Transcribe model, maybe even just an integration to Whisp
4.
by akreal 6mo ago
I also expected screenshots there, especially given the word "interface". Turns out, it's not about user interface (UI), it's about programming interface (kinda API). It allows calling window-related functions on Macinto
5.
by akreal 7mo ago
Docling: https://github.com/docling-project/docling
6.
by akreal 10mo ago
There is a dolphin language model project from Google and Georgia Tech: https://blog.google/technology/ai/dolphingemma/
7.
by akreal 10mo ago
As far as I understand, the MMS TTS models are trained from scratch (section 7.1 of [1]), they do not employ any SSL models. So the OmniASR SSL models are not useful here. What might be interesting is the newly released OmniASR data, becaus
8.
by akreal 1y ago
How is/was the data written to disks? Something like rsync/netcat?
9.
by akreal 1y ago
Probably a mix-up with the recently released Huawei model: https://news.ycombinator.com/item?id=44441447
10.
by akreal 1y ago
What is the motivation to move startup from Canada to Germany, if I may ask?
11.
by akreal 1y ago
I would be interested to see this reproducibility study.
12.
by akreal 2y ago
Could this (or something similar) be found in public access/some libraries?
13.
by akreal 2y ago
Telegram?
14.
by akreal 2y ago
Doesn't moving air help evaporation anyhow?
15.
by akreal 2y ago
The infrastructure problem should also have a relatively cheap technical solution. I assume the major bottleneck is bureaucracy, which largely has no technical workarounds so far.
16.
by akreal 2y ago
It took the modern human brain quite a bit of observations to discover and formulate f=ma, it was originally published in 1687.
17.
by akreal 2y ago
I was surprised to see this headline, because I had been using Plasma 5 on Intel graphics for almost a decade without any issues. After reading the article, it turns out that this is about Wayland, and I used X11. Wayland was indeed quite l
18.
by akreal 2y ago
Cool! HF Transformers is great for prototyping and research, but should not an interactive tool like this be based on something more speed-focused, like llama.cpp? Any plans for languages beyond English?
19.
by akreal 3y ago
Maybe this: "top designer Mark Krebs, who worked in Google's satellite and aircraft division" https://www.reuters.com/article/us-spacex-starlink-insight-i...
20.
by akreal 3y ago
Maybe you are referring to Stasi documents: https://www.stasi-unterlagen-archiv.de/en/archives/the-recon... Edit: link to the English version.
21.
by akreal 3y ago
I was curious about energy efficiency and took two samples from the linked MLPerf GPT3 results. H100 seems about three times more efficient than Gaudi2. 256 Gaudi2 600W TDP: 256 * (442.578 / 60) * 0.6 = ~1133 kWh 512 H100 700W TD
22.
by akreal 3y ago
Can't they install a normal Linux?
23.
by akreal 3y ago
A shameless plug, my colleagues made a demo where you could create a virtually infinite amount of voices with some sort of control of how they sound like: https://huggingface.co/spaces/Flux9665/ThisSpeakerDoesNotEx
24.
by akreal 4y ago
Which prompt did you use for translation? I'd be curious to try it for my task too.
25.
by akreal 4y ago
What else do you store besides movies and audio so that it becomes noticeble for you? I also use btrfs with zstd, but mainly see the difference for OS/apps files. Movies and audio don't compress as you noted, and code and scripts
26.
by akreal 4y ago
Not true for macOS.
27.
by akreal 4y ago
One could outsource the work to other people.
28.
by akreal 5y ago
There is also a comparison with CLD 2 in the repo of sister Python library: https://github.com/pemistahl/lingua-py#4-how-good-is-it CDL 2 seems to be slightly less accurate than CLD 3 on average.
29.
by akreal 5y ago
Redirect it. Double Asteroid Redirection Test mission is going to test soon if this would work in practice: https://en.m.wikipedia.org/wiki/Double_Asteroid_Redirection_...
30.
by akreal 5y ago
> 118 degrees Fahrenheit (48 degrees Celsius)

More ›