惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

Google DeepMind News
Google DeepMind News
aimingoo的专栏
aimingoo的专栏
爱范儿
爱范儿
D
Docker
I
InfoQ
Microsoft Security Blog
Microsoft Security Blog
G
Google Developers Blog
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
Vercel News
Vercel News
H
Hackread – Cybersecurity News, Data Breaches, AI and More
T
Tailwind CSS Blog
D
DataBreaches.Net
月光博客
月光博客
N
Netflix TechBlog - Medium
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
V
Visual Studio Blog
MyScale Blog
MyScale Blog
B
Blog
阮一峰的网络日志
阮一峰的网络日志
L
LangChain Blog
Recent Announcements
Recent Announcements
Microsoft Azure Blog
Microsoft Azure Blog
WordPress大学
WordPress大学

Show HN

Show HN: AI agents for UK GDAD PCF roles and their skills The Two Pillars: Mixer Mode and Meta-Software in the Reorganization of Software Work After AI GitHub - JaiCode08/teleport-env What 1,000+ Harness Experiments Taught Me About Self-Improving Agents Show HN: Liiists, a Markdown-first, iOS and CLI list app SwiperTab – Get this Extension for 🦊 Firefox (en-US) GitHub - kouhxp/fftext: Summarize, explain, fact-check, or translate any text, URL, or file. No GPU. No cloud. One command GitHub - sweetpad-dev/sweetpad: Develop Swift/iOS projects using VSCode GitHub - dogmaticdev/IRON: IRON a.k.a. Intermediate Representation Object Notation is a Interpreter/Database that is used to create Programming Languages. GitHub - sjhalani7/vaen: Package your AI coding harness into a portable .agent file, and share it across repos, teams, & the community without ever having to copy-paste instructions, skills, MCP config, or secrets. Show HN: Gandalf the Grader Show HN: Citadeld – replay any CI failure locally from a single file GitHub - tdortman/cuSBF: High-Performance GPU Super Bloom Filter coral-ai/claude-code-token-xray at main · Coral-Bricks-AI/coral-ai GitHub - ulyssestenn/funes: Funes is a Git-based framework for LLM-managed knowledge work: an AI Librarian ingests raw sources, builds an interlinked Markdown knowledge base, and uses it to produce cited reports, analyses, and other outputs. GitHub - ThatXliner/gah: Git Add Hunk, built for agents to use GitHub - harmont-dev/harmont-cli: Command-line client for the Harmont CI platform GitHub - brooksmcmillin/mcp-authflow: OAuth 2.0 Authorization Server framework for MCP servers GitHub - javaid-codes/audit-supply-chain-agents GitHub - amorey/gochan: A small library of common channel architectures for Go, inspired by Rust GitHub - arifozgun/OpenGem: Free, Open-Source AI API Gateway with Gemini, OpenAI & Anthropic Compatibility in 1 file GitHub - Pranesh950/BioPetals: 🌸 Run BIOxAI models at home, BitTorrent-style. Fine-tuning and inference up to 10x faster than offloading GitHub - cnguyen14/bounty-doctor: Diagnose a GitHub bounty issue before you waste hours: detects honeypot scam repos, AI-bot attempt swarms, and stale contests. Show HN: CoreMCP – MCP Server for On-Prem DBs Show HN: KittyHTML – Render HTML/CSS as an inline image in your terminal GitHub - bingud/filemat: Web-based file manager Show HN: TruthLens – Free multi-signal deepfake image detector GitHub - apexlocal-jz/claude-usage-tray: Windows system-tray app showing your Claude Code rate-limit usage at a glance. Zero deps, ~300 lines of PowerShell. Cross-IDE (works regardless of VS Code, Cursor, plain terminal). Release v0.1.2.1 · kouhxp/yapsnap GitHub - noopolis/moltnet: Self-hostable chat network for AI agents. Pre-built bridges for Claude Code, Codex, and the Claws. Rooms, DMs, history. No Slack bots, no Matrix, no glue code.
Speechmark — Private, on-device AI meeting notes for Mac
theone · 2026-06-19 · via Show HN

For macOS · Private · Public beta

Private meeting notes —
memory your own AI can read.

speechmark sits in your menu bar, turns each meeting into a clean note — decisions, action items, the words that mattered — and keeps it on your Mac, where Claude Desktop and other MCP-aware AI can read and search your past meetings on demand.

Free during the public beta · $79 $49 at launch·macOS 14.2 Sonoma+·Apple silicon·Recordings stay on your Mac

How it works

Three things, done well. Nothing else.

01 · Capture

It lives in your menu bar.

One click to record. The mic, the meeting audio, both — speechmark handles the routing. No browser extension, no bot joining the call.

02 · Listen

It hears who said what.

Speakers are separated automatically and labelled the first time you confirm a name. After that, every transcript reads like a script.

§

03 · Write

It writes the note you would have written.

Decisions, action items, and a brief recap — in plain editorial prose. Pick the model that fits your privacy bar: Apple Intelligence, a local Ollama model, or your own OpenAI / Anthropic key. Click any line in the note to jump to the exact moment it was said.

Ask your meetings anything.

Connect Speechmark to Claude Desktop and your transcript folder becomes a memory your own AI can reason over — privately, on your device. No competitor can hand you this, because their notes live in their cloud. Yours live on your Mac.

01

One click to connect. Settings → Assistant → Add to Claude Desktop. It installs as a Claude extension — no config files, no terminal, no API key.

02

Grounded in your meetings. Claude answers from your own notes — decisions, action items, attendees — and cites the meeting it came from.

03

Stays on your Mac. The connector reads your notes locally and makes no network calls of its own. You decide what any question surfaces.

speechmark · connected to Claude

What did we decide about pricing across my Acme calls?

Across three Acme calls you landed on annual billing at a 15% discount, with the per-seat floor still open. Two follow-ups are unresolved:· Send Acme the revised annual quote — you
· Confirm the discount with finance — Dana
from: Acme pricing review · Acme onboarding kickoff

Why Speechmark

The private alternative to cloud recorders.

Audio stays on your Mac

On-device transcription

No bot joins your call

BYOK — use your own AI key

Competitor data reflects their cloud-first defaults. Features may vary by plan.

I used to leave meetings with three pages of bullet points and no idea what we'd actually agreed. Now I leave with a paragraph that reads like a memo from someone who took it seriously.

Mira K.

Director of product · early access

Who it's for

Built for people whose meetings matter.

01 / 04

Capture decisions and action items — not a wall of bullet points your team won't read.

Speechmark tracks who said what and when, so your sprint retros, PRD updates, and stakeholder readouts practically write themselves. Stay in the conversation instead of racing to keep notes.

Privacy

Your meetings stay on your machine.

Recordings and transcripts stay on your Mac. The only things that can ever leave are ones you switch on yourself — a cloud AI model (transcript text only) or anonymous usage stats. Both are off by default, and you control them.

Pick the AI model that fits your privacy bar — Apple Foundation Models (on-device, default), a local Ollama model (fully on-device, no API key), or your own OpenAI / Anthropic key — only the transcript text is sent, never the audio.

  • Recordings & transcripts stay on your Mac
  • Transcription runs on-device
  • No bot dialed into your call
  • No account, no login, no sign-up
  • Usage stats: opt-in, off by default, yours to control
  • BYOK cloud model — only transcript text is sent

Read the full privacy policy

Questions

Common questions, answered plainly.

Does Speechmark upload my meeting audio to the cloud?

No. Recording and transcription happen entirely on your Mac, and the audio never leaves the device. If you pick a cloud AI model for the summary (OpenAI or Anthropic, with your own key), only the transcript text is sent — never the audio. You can also stay fully on-device with Apple Foundation Models or a local Ollama model.

How is it different from Otter.ai, Fireflies, or Granola?

Those are cloud-first: your audio is uploaded to their servers and you pay a recurring subscription. Speechmark keeps audio and transcription on your Mac, needs no account, doesn't send a bot into your call, and is a one-time purchase instead of a subscription.

Does a bot join my call to record it?

No. Speechmark records the microphone and meeting audio directly from your Mac’s menu bar. Nothing joins the meeting, and other participants see no recording bot.

How much does it cost?

Speechmark is a one-time purchase of $79 per Mac, with no subscription. It’s currently a free public beta you can download and run today; buying now locks in a founding-customer price below the launch price.

What do I need to run it?

macOS 14.2 (Sonoma) or later on an Apple silicon Mac.

Can I ask my meetings questions with Claude?

Yes. Speechmark installs a local connector for Claude Desktop in one click — no API key, no config. Claude then answers from your own meeting notes and cites which meeting each answer came from, while the notes stay on your Mac.

Try it on your next meeting.

Free during the public beta. No account, no credit card, no waitlist — download and run it right now.

macOS 14.2+

It’s an early build, and your feedback shapes it. Found a bug or have a request? Send feedback →