christiansafka

That is a good looking homemade burger

Server-side yes - the app ships with its own open-source remote daemon that gets setup when you connect to the server

A native Mac app as a modern tmux replacement. I manage agents and terminal sessions across several GPU machines and having it all in one UI has been great for reducing cognitive load. If anyone's interested in being an early adopter, we're at https://haventerminal.com

If anyone is looking for a slightly higher level management of persistent sessions, we just launched https://haventerminal.com for Mac

I would also describe haventerminal.com as making tmux pretty and usable! but without any setup steps, first class support for agents, and management of multiple machines. Disclaimer: It's a macOS app we just launched and it ships its own open source session persistence.

We've published our learnings and results as part of our entry to the Interspeech 2026 Audio Encoder challenge. Code and model will be released on the same url

Found a post from Twilio using the same tech stack in 2013! What a trip. Cool project!

https://news.ycombinator.com/item?id=5409914

Over the weekend I built a website for teachers and parents to ask questions from Maria Montessori but couldn't fit all the relevant docs into context windows. The other RAG libraries were either doing too much (IMO) or involved connecting to an external Postgres which felt extra for what I was building. So I created this minimal RAG plugin, built for side projects. It is actually unrelated to nextjs, should run in any nodeJS env.

If anyone finds this useful I'll probably add some parsing for non-text formats like PDF, images, videos.

Repo: https://github.com/christiansafka/nextjs-rag

Thanks for this feedback! We're rolling out an update soon that will allow you to hear your translated voice without joining a call. CPU usage is on the radar, hope to make some progress there soon as well.

Yes, we're planning a Windows launch as well! If you need something today, we also have our own video conferencing platform with AI interpretation built in that works in-browser. It's at https://startpinch.com/meeting

Our first focus is desktop, but we'd love to get a mobile app out (or let someone else build one with our API)

Yes, if you're watching TV on your computer the app should translate all the speech. We haven't done any source separation work (musical singing vs actual speech) so if there's music it may pick up some of the lyrics.

Thanks! We're planning to add support for a lot more languages over the next two weeks.

Fun idea, are you thinking about a language learning use case?

Yes, right now we're ranging from 0.75-3 seconds for the translation to start, and we're hoping to move the average time lower with our next updates. There will always be some limitation to how fast we can translate (different languages have different sentence structures and phrasing), but for livestreaming usually you'd have even a bit more wiggle room for the latency.

Also in case you're interested in the logistics of using us for livestreaming: If our current platform won't work for your use-case and you need to use OBS + a virtual camera, it's on our roadmap.

We're hoping companies with international teams will switch over fully (we have internally), but our initial goal is to attract a subset of the market that has cross-lingual needs and unblock them as much as possible from using it more.

Fantastic question. Our opinion on this is that the higher-bandwidth we can make the communication, the more useful it will be. The reason we've moved from IRC->VoIP->Video is because of the efficiency of information transfer and additionally the empathic element of face-to-face conversation.

From the technical side, speech to speech models have more potential for accuracy (no explicit ASR, no audio->text information loss). We have a few options on mimic'ing nonverbal elements - we could decide when to naturally mix in the original audio, or train our end to end model to handle those nonverbal audio chunks. We'll be trying both but likely the first option on the sooner side!

Thanks, there are certainly a lot of fun and challenging problems to take on in the space. On scaling, the agent architecture isn't limited to one machine, so you can also autoscale your machines. It's essentially python's Celery if you've tried that. It gets more tricky when you require GPUs though!

It's hard, don't do it :D We have a few supply chain companies trying us out though! Would love to hear more about your experience.

Appreciate the feedback. On the video side, we currently synchronize it to play out with the translated audio (as often as possible), matching when you started speaking to the moment the translated audio starts. Mentioned in another comment but we're still working on audio mimicking (voice clone then inflection transfer). Our model does a lot that Google Translate doesn't, even just around translation, such as taking into account who you're talking to in the meeting and the conversation context. + we have to do it much faster, so smaller audio chunks at a time!

Yes! LiveKit is great - and we are using livekit agents but had to override a few low-level library components for our use case.

I didn't make this clear enough in the post, but we're still working on voice cloning and inflection transfer. Voice cloning is easier, but to support inflection transfer we have to modality-align an LLM.

Thanks! We have a virtual camera on our roadmap as well, but by building the conferencing platform end to end we can optimize both latency and conversation UX to a much higher degree. We're also lucky to be building this now and not five years ago - there are some solid webrtc infra companies and open source projects to build on.

We noticed that Swedish isn't currently working properly, but we weren't able to replicate this with any other languages. Please let us know if it's still having issues!