Skip to content
AI NEWS JOURNAL
AI, IN PERSPECTIVE

Search AI news

Search headlines, summaries and articles to connect the dots.

Newest first

Show HN: Apron Chat Protocol

Article URL: https://apron.chat/ Comments URL: https://news.ycombinator.com/item?id=50470343 Points: 6 # Comments: 1

Show HN: RealtimeVAD – trading false positives for 110ms earlier end-of-speech

Hey guys. I recently watched the latency reports for my local voice assistant, which show a waveform plus silero probability for voice activity. I noticed silero lags visibly on the speech off signal.There had to be a method to get this signal closer to what I felt is optimal. My idea was to use a 10 ms energy detector with a threshold that adapts to environment or room noise. Which worked amazingly well. It created more false positives but in my speculative generation pipeline that would not matter. If it's wrong and the user continues speaking I can abort the pipeline fast.I benchmarked my solution over ~3200 turns from my assistant logs and it came out ~110 ms faster initially. Still 100 ms now after some a fix which was needed for quiet speech.I made it public if you want to have a look. If you measure it on your data I really appreciate any feedback. There's a benchmark script in the repo.Thx for reading, have a good day folks. Comments URL: https://news.ycombinator.com/item?id=50474586 Points: 2 # Comments: 0

Show HN: TempCD – Temporarily CD in a directory with one command

TempCD is just a small thing I made, its only purpose it to eliminate having to enter cd again after only wanting to do one command in a directory. Super simple, just:> tcd ~/example ~/example> command command output >That's it, just returns you back to where you were before without having to type cd again. Comments URL: https://news.ycombinator.com/item?id=50477500 Points: 1 # Comments: 0

v1.0.0

The llama.cpp project has released version 1.0.0, marking a significant milestone with numerous improvements across multiple platforms including CUDA, Vulkan, Metal, SYCL, OpenCL, and WebGPU. Key updates include enhanced support for quantized models (Q5_0, Q5_1, FP8), optimizations for flash attention and MoE caching, improved server security features like CORS protection and SSRF mitigation, UI enhancements such as sidebar reorganization and live-preview markdown editing, and better cross-platform compatibility. The release also includes bug fixes, CI/CD improvements, and updated dependencies.

Clipto enables ChatGPT and Claude to search and analyze years of people’s meetings and video calls

Artificial intelligence assistants such as ChatGPT and Claude have become essential tools for millions of workers worldwide, helping them compose emails in seconds, generate complex code and draft reports by looking at whatever data that person has access to. But there’s one thing that’s missing from those chatbots’ knowledge. Ask them what was discussed in […] The post Clipto enables ChatGPT and Claude to search and analyze years of people’s meetings and video calls appeared first on SiliconANGLE.

Hatz AI raises $24M to help small businesses go AI-native through their MSPs

Large enterprises have emerged as the foremost adopters of artificial intelligence technology, and it’s giving them an unfair advantage over their smaller competitors. Looking to change that is a startup called Hatz AI Inc., which has just raised $24 million in early-stage funding to empower managed services providers with the tools they need to keep […] The post Hatz AI raises $24M to help small businesses go AI-native through their MSPs appeared first on SiliconANGLE.

Local Analytics Doesn’t Need a Database Server

From interactive maps to graph exploration, DuckDB makes it possible to analyze data without a database server. Ilya Boyandin shares how this approach led to SQLRooms, a toolkit for building analytics applications that bring data, visualizations, and AI-assisted analysis together on the user’s machine.

Qvyon launches with $8M to improve safety for physical AI deployments

Physical AI validation infrastructure and software development company Qvyon Technologies Inc. launched today with $8 million in total funding to help companies validate autonomous vehicles and robots before deployment. MaC Venture Capital led the funding, with participation from BDC Capital, BKR Capital and Spotlight VP. Physical AI is artificial intelligence that moves from digital brains […] The post Qvyon launches with $8M to improve safety for physical AI deployments appeared first on SiliconANGLE.

Show HN: Builderoom – find indie-made software by describing what you need

Hi HN. Builderoom is a site where indie devs, designers and makers list what they build, and you can look for stuff by just describing what you need, like "physics simulation running in the browser". Browsing and normal keyword search work without an account.Why I made it: a lot of the time I know what I need but not what the thing is called, so keyword search doesn't help much. And whenever I ship something, every launch site wants the same long form again. Comments URL: https://news.ycombinator.com/item?id=50464162 Points: 1 # Comments: 0

Show HN: Lynx – open-source fully-customizable AI workspace

I was finding that managing multiple agent threads with walls of text was getting tiring and inefficient for me and that I wanted a more visual way to interact with my agents, so I built Lynx.Lynx lets you quickly build fully custom interfaces connected to your data sources (e.g. via MCP or API) and trigger agents running on your machine from them (e.g. with Claude Code or Codex).I built it as an experiment but have found myself doing 80-90% of my work from inside Lynx. I have my own Conductor-like workspace for coding, a Mail client, a workspace for my writing, and various others for dashboards and explorations.A few things are really cool about it:- Adding buttons to a UI that trigger a Claude Code agent running on my machine (e.g. to review a PR, get started on a task from Linear, or look for typos in my text)- Iterating on the interfaces is really easy, particularly because of a point tool that let's you select what you want changed and an agent will update it live- Switching between workspaces takes a fraction of a second (much faster than e.g. switching between websites)- It's all local :)Harnesses have been moving towards native support for artifacts and MCP-UI which kind of go in this direction but I think the current models add to context fragmentation rather than actually help it, so I flipped the model on its head and built an AI workspace / harness where interactive interfaces are at the core.Calling this an AI workspace for now but honestly unsure what the proper term should be. Is this a visual harness? A hub for personal software?Keen to hear feedback about the conceptual model and what others have done that might be similar! Comments URL: https://news.ycombinator.com/item?id=50461775 Points: 1 # Comments: 0

The Definitive Fine-Tuning Guide for 3D Generation Models

This blog post provides scripts and Docker containers for fine-tuning TRELLIS.2, a 3D generation model, using LoRA adapters on footwear data from the Google Scanned Objects (GSO) dataset. The approach focuses on improving geometric accuracy by training only the sparse-structure and shape flow diffusion transformers while freezing the texture stage. The process includes preprocessing with Blender to generate 16 views per object, converting meshes into latents, and training with Hugging Face Jobs or local Docker. Results show significant improvements in F-score, chamfer distance, and normal consistency on held-out test objects, demonstrating effective adaptation for specific 3D object categories.

Adobe’s AI experiments let you walk around in photos and interview your characters

Some of the most impressive tech that Adobe showcases at its annual Max conference each year is the experimental features it's developing, and the latest of those tools are being announced today: they can transform images into explorable 3D scenes, design AI characters by talking to them, and turn photo editing into a copy/paste process. […]

Show HN: CodeLens AI-Augmented Code Review Tool

As AI writes more code, the bottleneck has become human-driven code reviews. Code Lens serves to aid that task by augmenting the review outside of line-level AI feedback. Check it out and let me know what you think. Thanks.Try it on a public PR: npx @blong8334/codelens pr https://github.com/expressjs/express/pull/7574 Comments URL: https://news.ycombinator.com/item?id=50446651 Points: 1 # Comments: 0