Firefox should integrate that in their Reader Mode (the default System Voices are often very un-listable). Would seems like an easy win, and it's a non-AI feature so not polarising.
Not sure about macOS or Windows, but on Linux Firefox uses speech-dispatcher, which is a server, and Firefox is the client. Speech-dispatcher then delegates the text to the correct TTS backend. It basically runs a shell command, either sending the text to a TTS HTTP server using curl, or piping it to the standard input of a TTS binary.
Speech-dispatcher commonly uses espeak-ng, which sounds robotic but is reportedly better for visually impaired users, because at higher speeds it is still intelligible. This allows visually impaired users to hear UI labels more quickly. For non visually impaired users, we generally want natural sounding voices and to use TTS in the same way we would listen to podcasts or a bedtime story.
With this system, users are in full control and can swap TTS models easily. If a model is shipped and, two weeks later, a smaller, newer, or better one appears, their work would become obsolete very quickly.
> Documenting it, however, was a marathon that spanned many weeks of focused work.
I like how detailed they were in their writing (even though I was suspecting the graph function to be the culprit all along) ! Big Kudos though to get to the ground of it and writing it all down for others to learn ! Also, indeed this would make a great template for LLM problem-decomposition / solving.
> All that remained was to decide what to do with my life. From a spiritual perspective, there are only two career paths one can take: farmer or artisan. Anything else unavoidably involves doing evil or is essentially meaningless.
Seems shocking at first, but the more you think about what our SWE works does, for whom, and who benefits the most of it ... IMO it makes sense.
It sounds more like a depression and stress brained reduction to me. Tends to put you in a very binary and extreme thinking mode in my lived experience.
Also I inherently disagree with the idea of meaninglessness the author presents there. Meaning is relative to man, man makes meaning. There is no objective meaning and so you have to choose it for yourself.
For many, software engineering is an artisan endeavor (hence why many are freaking out over AI, it removes their enjoyment of the process even as others, who are solution oriented, like the final output and what problems it can solve without giving a shit about the code, two different types of people).
You can apply software development skills to public good. It's just not the most common path, nor the best-paid one. I should also highlight that SWE has one of the most prolific gift economies out there.
Well, "essentially meaningless" does away with basically anything that isn't water and food, so lets be measured. Working on video games could be done in an ethical, sustainable and non-evil way, but also one can argue is "essentially meaningless" together with everything else too, including "artisan".
The response timing in the chart in the blog post shows that even with perfect precision/recall Sparrow-1 also has the fastest true positive response times.
The turn taking models were evaluated in a controlled environment with no additional cascaded steps: LLM, TTS, Phx. This matters to get apples to apples comparison: without the rest of the pipeline variability influencing the measurements.
The video conversation examples are sparrow-1 within the full pipeline. These responses aren’t as fast as sparrow itself because the LLM, TTS, facial rendering, and network transport also take time. Without Sparrow-1 they would be slower. Sparrow-1 enables the responses being as fast as they are, and with a faster CVI pipeline configuration the responses can be as fast as 430ms in my testing.
If you watch the demo video you can see how they would get this: the model is not aggressive enough. While it doesn't cut you off, which is nice, it also always waits an uncanny amount of time to chime in.
Great post! Indeed, it s deeply disappointing to see how both the tech industry and scientific community have fallen into the same attention-seeking trap: hyping their work with vague, sensational claims, only to later "clarify" with far more grounded—and often mundane—statements.
This tactic mirrors the strategies of tabloids, demagogues, and social media’s for-profit engagement playbook (think Zuckerberg, Musk, and the like). It’s a race to the bottom, eroding public trust and undermining the foundations of our society - all for short-term personal gain.
What’s even more disheartening is how this dynamic rewards self-promotion over substance. Today’s "experts" are often those who excel at marketing themselves, while the most knowledgeable and honest voices remain in the shadows. Their "flaw"? Refusing to sacrifice integrity for attention.
The new CEO centered AI ("It's Time to Evolve Firefox Into an AI Browser") in his first communication to the community. Spawned at least three new forks and introduced people to LibreWolf.
His first communication reduced trust: "It is a privilege to lead an organization with a long history of standing up for people and building technology that puts them first."
Now let's put people first by making Firefox an AI first browser. Enzor-Demeo would have made an excellent Microsoft product manager. Too bad he didn't get the job.
Vibekit is what I would thus far deem the best automatic, yet mostly overlooked, sandboxing solution for agentic LLMs.
It delivers a full-featured sandbox that seamlessly integrates with most LLM providers and server/local sandboxes - it works out-of-the-box while literally keeping your agent in-the-box.
I reviewed the mini-blog post and initially thought: "Okay, this doesn't seem unreasonable". Then I clicked over to the "About" section, only to find out the author is the CTO of Meta (and proudly at Facebook for two decades).
Then took a closer look at the latest post, "Love what you do." Really? If "loving what you do" means contributing to Facebook/Meta’s legacy of facilitating genocides, exploiting users, running unethical social experiments, and overall polarizing societies to the brink of destruction just for profit - then your "life advice" is just hollow, superficial nonsense. Screw you, "Boz" - we don’t need that kind of hypocrisy at HN.
I had the same thought, how can you continue working for Meta if the leader happily undermines democracy for profit and enjoys schmoozing with the current administration who have no scruples of dismantling our democratic institutions and world order.
I get not everyone can leave a company if their life depends on it and they have to support a family, especially in this market.
But this guy is probably a millionaire already. He's got the luxury of working for more world positive companies or projects.
But him choosing to continue to work for Zuck sends a clear signal what his values are.
It's all just self embellishment and rationalisation with these guys for the horrible stuff they did. Even if they think its genuine, this Philip K Dick quote fits exactly "Many men talk like philosophers and live like fools".
reply