Cicada runs on macOS and captures system audio to produce bilingual subtitles, real-time translation, and structured meeting notes. It works with Zoom, Teams, YouTube and other streams without sending audio to the cloud.
Why it matters. Anyone who spends time in online meetings or watches live streams can get instant captions, AI-generated summaries, and even speak translated audio back into the call.
Try it:Download the latest release from GitHub and drag the app into /Applications.
spotted 2 h ago
Today in one paragraph. Today's theme is about cramming heavyweight platforms into tiny interfaces, like a full macOS virtual machine that lives inside your terminal. From disc-burning AI to unattended CLI agents, the new tools let you run, edit, or evaluate heavyweight workloads from a single command line or browser tab.
Recent advances in interactive world models have made it possible to generate controllable, dynamic environments from action inputs. Yet current models struggle to maintain coherence beyond roughly one minute of rollout, a limitation rooted not in model capacity but in the scarcity of long-horizon t
Reasoning vision-language-action (VLA) models pair physical control with natural-language rationales that supervisors can read to monitor embodied safety. Prior attacks on VLA reasoning corrupt the rationale itself and leave visible evidence in the text. We show that visual perturbations can instead
1 upvotes
The conversation
What builders are arguing about on Hacker News, top 2 of 3
Anthropic released a system that generates zero-knowledge SNARK proofs showing each answer was produced by a specific model. The proof can be verified without revealing the model’s weights or the prompt.