Voice-first AI workspace for meeting, notes, dictation and voice skills. Works offline. Try it at: https://teapaw.org/t.co/DZaE3f0Soa iOS App: https://teapaw.org/t.co/HNI3kgtcMq
Palo Alto
Joined August 2024
- Tweets126
- Following22
- Followers152
- Likes114
Mumble Note 0.8.8 is out.
Ask Chat a question across your notes and it tells you which notes the answer came from.
Select any sentence, tap once, it's a to-do.
Get a nudge when a long recording is written up.
Also: Starter now includes 10 AI chats a month, and Apple Watch recordings sync reliably even on a bad connection.
You fix the same misspelled name every week, and every week it comes back.
That's not an accuracy problem. It's a memory problem. Nothing in the audio says whether it's Sean or Shawn, so the tool has to already know which one you mean.
Mumble Dictation keeps a personal dictionary on your Mac. It picks up the names and terms you actually use from your own typing. Fix a spelling once and it still holds next month.
cmd+space is fast. it just needs your hands on the keyboard.
Open mode extends that reach to the browser.
Try Fn +6 ：mumble the question out loud and the results are up
1
3
109
Mumble Dictation is Mac only. 60 minutes of full AI processing free to try.
heymumble.com/dictation?utm_…
47
$144 a year to type, and it still stops working on a plane.
That's the deal with most dictation tools right now. Cloud only,so every sentence takes a round trip to someone else's server before it lands in your doc.
We went the other way. Mumble transcribes on device by default,
So offline is just a normal tuesday. $50 once, and you keep it.
wifi off, just mumble it~
This made our day, thank you! Glad the clipboard edit trick is landing for you！
Really appreciate the mobile app suggestion, we'll definitely be thinking about it!
And if you ever run into any issues using the app, just mumble to us anytime 😂
I've used Wispr for a good chunk of my productivity, but recently I switched to @mumblenote for my Mac, and I'm not looking back.
One thing I really love: I can just mumble out my thoughts (😂 pun humour), then edit them right there in the clipboard area before dropping them into the chat. No hunting for anything. And that's just one of the many highlights!
The dictation model downloads right onto your Mac (~2GB) and runs entirely offline, so nothing gets sent to the cloud.
I'm very curious to see where they take it, and really hoping for a mobile app in the future.
A recent study had MBA students debate an AI ,some by voice, some by typing.
The voice group explored measurably more diverse ideas. Researchers think it's because speaking lowers the cognitive cost of generating thoughts: you don't self-edit mid-sentence the way you do at a keyboard.
For those of you whose natural habitat is a terminal window. Typed prompts come out compressed and under-specified. Spoken ones ramble — and the rambling is where the context lives.
(Also, Stanford measured speech at 3x typing speed. So you're exploring more ideas, faster.)
Just mumble to your Mac, and let your words take you there.
1
62
I haven't typed to Claude Code in 30 days.🤯
"Refactor the auth middleware, run npm run dev, check if the Supabase client still works"
Mumble dictation gets every word right.Even file names！
No fixing, no re-typing.
It builds a personal dictionary from what you actually talked automatically. The more you use it, the better it knows your stack.
Hands off the keyboard. just mumble it 👇
heymumble.com/dictation
116
Speak freely. Stay private.
Mumble Dictation is live today.🚀
local - first, top-tier accuracy
→ Prompt at speaking speed, in Cursor, Claude, anywhere
→ 40+ languages, and it doesn't break when you mix them mid-sentence
→ Works offline, even on the plane
just mumble it 👇
heymumble.com/dictation
We're on Hacker News today！🎉
Shared how we're solving two things that have bugged us about dictation apps on the market: fake privacy (most still ship your audio to the cloud) and generic ASR that's terrible at your actual vocabulary.
Mumble is built to be the most private, and the most fluent in you.
Meet us on HN.news.ycombinator.com/item?id…
Mumble AI retweeted
THIS DEVELOPER JUST KILLED THE VOICE CLONING INDUSTRY WITH ONE GITHUB REPO.
Claude can now speak in your own voice across 23 languages through a single MCP call.
Free. Local. Offline.
What used to cost $22 to $330/month with ElevenLabs is now open source under the MIT license.
Jamie Pine just dropped voicebox on GitHub, and it already has over 41k stars.
It clones any voice from just 10 seconds of audio, lets you choose from seven different TTS engines, and runs entirely on your own machine, so your voice and recordings never leave your computer.
Connect it to Claude Code, Cursor, Cline, or any MCP compatible agent with a single voicebox.speak call, and your AI can start replying in your cloned voice.
It also adds global voice dictation that works in any app with one hotkey and can generate speech in 23 languages, including Arabic, Japanese, and Polish.
One open source app now replaces both ElevenLabs and Wispr Flow.
If you’ve been waiting to give your AI agent a voice, this is it.
Repo below 👇
Mumble 1.1 is out.
Four things that make voice a real input method on your Mac:
1. Meetings: edit attendees, translate summaries, pick templates while recording
Add or remove attendees anytime. Pick speakers from the live transcript. Regenerate summaries in any language. Pick output templates right from the recorder.
