62 karma · joined May 6, 2019
Not for everyone obviously. But there are many, many, many, options at this level of quality.
I am not claiming these Sennheisers are the best. The parent claimed the Airpods are the best. Which is not true. That's all I'm saying.
Top-shelf wireless earbuds aren't from Apple. Same for smart watches.
One concrete example I'm sure these Skyrim mods aren't using is: enums in structured outputs [1] with a finite list of locations/characters/topics/etc that are allowed to be discussed. The AI is not allowed to respond with anything that is not in the enum. So you can give it a list of all the locations in the game in a huge array and it would be forced to pick one.
[1] https://platform.openai.com/docs/guides/structured-outputs#a...
I think the thing you brought up at the beginning is the most practical path forward, someone with the technical know-how and business acumen needs to start a company. Apple and Google are quite weak now, and there are lessons to be learned from the Librem 5 and PinePhone. If enough people try, someone will eventually break through.
Unsolicited advice: Stop using the Github web UI asap. It will slow you down. VScode has git integration. There's an upfront cost to learning it but it's worth it.
If you're willing to take the plunge it's best to use the git command-line interface directly. It pays in the long run. But I understand it's kind of daunting at first.
Whatever you decide, whether its VScode or the git CLI, just get off of the Github web UI. Trust me. I look at my old projects from when I was in high school and I deeply regret that I didn't do version control. So much information that gets lost.
To be honest this librebox project looks kinda fishy to me. Seems like the author never used version control before. They put everything into one commit [1] and are making changes to the repo using the Github web UI. I wish the dev the best of luck but I wouldn't be surprised if this project gets abandoned in a few months
[1] https://github.com/librebox-devs/librebox-demo/commit/e70ea3...
Other than for the "live transcription" usecase (that they made unnecessarily complicated), I don't see how this is any better than running Whisper.cpp directly. Other people in this thread are basically saying "ffmpeg's interface is better understood" [2] but LLMs make that point moot since you can just ask them to do the drudgery for you.
[1] https://medium.com/@vpalmisano/run-whisper-audio-transcripti...
Of course you could disagree with my prediction and that these big tech companies are going to build MASSIVE gpu farms the size of the Tesla Gigafactory which can run godlike AI where nobody can compete, but if we get to that point I feel like we will have bigger problems than "AI react code is better than AI solidjs code"