frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Show HN: VibeWhisper – macOS voice-to-text with push-to-talk,cloud or 100% local

https://vibewhisper.dev/
2•AleksDoesCode•1h ago
Hey HN,

I built VibeWhisper because I was using Claude Code all day and the bottleneck wasn't thinking — it was typing. I tried an existing voice-to-text tool, loved the workflow, then saw the price: $20/month for a thin wrapper around OpenAI Whisper where 10 minutes of API usage costs $0.06. So I built my own in a weekend.

How it works: Hold a configurable global hotkey (default: Right Command), speak, release. Transcribed text appears in whatever text field has focus — VS Code, Terminal, Slack, browser, anything. No clipboard hijacking — it uses the macOS Accessibility API to inject text directly at the cursor position (same APIs screen readers use).

Two transcription engines:

Cloud: OpenAI Whisper API via your own key (~$0.006/min, fastest and most accurate)

Local: WhisperKit running on-device via CoreML/Apple Neural Engine. 10 models from Tiny (50 MB) to Large V3 Turbo (1.6 GB). Your voice never leaves your Mac. No API key needed, no internet needed. The app auto-recommends the best local model for your hardware.

Tech details for the curious:

Native Swift/SwiftUI, no Electron. Uses less RAM than a Chrome tab.

CGEvent taps for global keyboard hooks

Accessibility API for text injection

API key stored in macOS Keychain

Universal binary (arm64 + x86_64), macOS 14+

No backend, no accounts, no telemetry

Pricing: $19 one-time. No subscription. 15-minute free trial (actual recording time, not calendar time), no credit card, no signup. If you use cloud mode, you pay OpenAI directly at their rate. If you use local mode, it costs nothing after purchase.

I use it myself 2-3 hours a day. My main use case is dictating long prompts to AI coding assistants — I can speak a 400-word prompt in 45 seconds instead of typing it in 3-4 minutes. I also use it as a thinking tool: talking through problems out loud and capturing the output.

Website: https://vibewhisper.dev

Happy to answer any technical questions.

Comments

yarivk•1h ago
Nice project.

Voice to text with a push to talk workflow makes a lot of sense for developers who don't want constant background listening.

Curious how the local mode performs compared to the cloud version in terms of latency and accuracy

Show HN: FileShot – zero-knowledge file sharing, nothing locked behind accounts

https://fileshot.io/
1•GraysoftDev•32s ago•0 comments

Google's Nano Banana 2 promises Flash speeds with Pro results

https://thenewstack.io/nano-banana-2/
1•blazercohen•2m ago•0 comments

Show HN: ReleasAI – Paste Git commits, get polished release notes in <1s

https://releasai.vercel.app/
1•pramhatibie•2m ago•0 comments

Show HN: I made a AI slides editor

https://www.ablo.finance/
1•lukasanderss•2m ago•0 comments

Agent Skills Marketplace

https://skillsmp.com/
1•pama•3m ago•0 comments

I don't know marketing, so I built a tool to track what I do

https://www.buildemotion.com/
1•Kamil_KKA•4m ago•1 comments

To update blobs or not to update blobs

https://codon.org.uk/~mjg59/blog/p/to-update-blobs-or-not-to-update-blobs/
1•trelane•4m ago•0 comments

I Spent $120 Trying to Make an AI Vertical Drama About Cats. It Was a Disaster

https://www.streaming-radar.com/p/i-spent-120-trying-to-make-an-ai
1•lbostral•5m ago•0 comments

Show HN: Construct Computer – Agentic Cloud OS for Daily Work

https://construct.computer
1•ankushKun•5m ago•0 comments

Finding Peaks with SIMD

https://halleysfifthinc.github.io/posts/find-peaks-simd/
1•allenofthehills•5m ago•0 comments

Building an Inference Engine in 1,800 Lines of C++

https://linuxtoaster.com/blog/toasted.html
1•dirk94018•5m ago•1 comments

$82,000 in 48 Hours from stolen Gemini API Key

https://old.reddit.com/r/googlecloud/comments/1reqtvi/82000_in_48_hours_from_stolen_gemini_api_ke...
1•virgildotcodes•6m ago•0 comments

So You Want to Work in Hardware

https://ourancientfuture.substack.com/p/so-you-want-to-work-in-hardware
1•ls65536•7m ago•0 comments

Aria-haspopup might not do what you think it does

https://www.matuzo.at/blog/2026/aria-haspopup-menu
1•speckx•7m ago•0 comments

OpenAI amends Pentagon deal as Sam Altman admits it looks 'sloppy'

https://www.theguardian.com/technology/2026/mar/03/openai-pentagon-ceo-sam-altman-chatgpt
1•TutleCpt•8m ago•0 comments

How to combat Claude's code duplication, ranked from least to most AI-native

https://ngof.nikhaldimann.com/p/4-ways-to-combat-claudes-code-duplication
1•nikhaldi•9m ago•0 comments

Show HN: Core Rth. A governed AI kernel for engineers who don't trust their LLMs

https://github.com/rthgit/CORE-RTH
1•christianrth•9m ago•0 comments

Show HN: DataPilot – SQL workspace with scheduling, and on-prem execution

https://getdatapilot.com/
1•vincenzo_nudo•10m ago•0 comments

Show HN: PDFBolt – HTML to PDF API with Git-like template versioning

https://pdfbolt.com/
1•Ametrin•11m ago•0 comments

Show HN: ScrollMind – Learn how AI works by scrolling a feed, not a textbook

https://scrollmind.ai/course/intro-ai
1•bkraszewski•11m ago•1 comments

Go code is leaving 90% of the CPU idle until now

https://samuelberthe.substack.com/p/your-go-code-is-leaving-90-of-the
1•samber•11m ago•0 comments

Pentagon's Anthropic Designation Won't Survive First Contact with Legal System

https://www.lawfaremedia.org/article/pentagon%27s-anthropic-designation-won%27t-survive-first-con...
3•elsewhen•13m ago•0 comments

Show HN: My API was leaking its full database schema. I found out by accident

https://github.com/brakit-ai/brakit
1•speak2aditya•14m ago•1 comments

Show HN: Vynix – Mobile app with 100 AI models for image/video/audio generation

https://www.vynix.app
1•cosmicmeta•14m ago•1 comments

Show HN: I built a serverless wrapper for the EU VIES with caching and webhooks

https://vatflow.net
1•QuiCreatDev•14m ago•1 comments

"bunnie" Huang prepares a dev board for Baochip's "mostly-open" X1 RISC-V MCU

https://www.hackster.io/news/andrew-bunnie-huang-prepares-the-dabao-a-dev-board-for-baochip-s-mos...
1•MassPikeMike•14m ago•0 comments

Anthropic's AI model Claude gets popularity boost after US Military feud

https://www.theguardian.com/technology/2026/mar/02/claude-anthropic-ai-pentagon
3•beardyw•15m ago•0 comments

Gimp 3.2 RC3: Third Release Candidate for Gimp 3.2

https://www.gimp.org/news/2026/03/02/gimp-3-2-RC3-released/
1•marcodiego•17m ago•0 comments

Tetris alleviates PTSD symptoms in nursing staff

https://www.thelancet.com/journals/lanpsy/article/PIIS2215-0366(25)00397-9/fulltext
1•doener•17m ago•0 comments

The New Postman Is Here: AI-Native and Built for the Agentic Era

https://blog.postman.com/new-postman-is-here
1•johnhamlin•21m ago•0 comments