
own your speech; open source; dictations, meeting notes and computer use all powered by on device models for macOS and iOS maintained by @pHequals7
on-device
Joined June 2026
- Tweets398
- Following233
- Followers203
- Likes995
muesli retweeted
🎙️ The meeting ends and the words are already gone. Muesli 0.8.6, released November 8, 2026, keeps a searchable screen history on this Mac, and screen capture stays off until you turn it on.
Meet Muesli by Pranav Hari. MIT License. Version 0.8.6, published November 8, 2026. A native macOS app for dictation and meeting transcription, built for Apple Silicon. Speak into any text field, and capture a meeting without a bot.
🕘 Revisit the day from this Mac
Time Turner lets you browse and search screen history, catch up with recaps, and pick up unfinished work with to-dos. Screen capture is off by default, and the history stays on this Mac. Time Turner recaps send sampled screen text and excerpts from other sources you include, such as saved meetings, to your selected Time Turner provider.
🎥 Show a coding assistant the screen
Vidshots records a window or a screen region while you explain it aloud, so a bug, a UI, or a change can go out as video, narration, and the captured picture. During a meeting, record a selected window or display with the meeting audio and play it back locally. That recording needs Screen Recording and Microphone permission, and the people in the meeting need to agree before you record.
⌨️ Take the next words from the keyboard
Cotypist offers on-device suggestions as you type. Accept one with Tab, or take it one word at a time with Option + Right Arrow. Turn it on in Settings → Experimental. Accepted Cotypist suggestions are saved in your local Timeline with preceding text and app name.
🗣️ Watch the sentence form
During hands-free dictation with supported Parakeet and Whisper models, words appear as you speak. Enabled cleanup runs when you stop. You can keep dictating in other apps while a meeting is still recording. MedASR, for English medical dictation, needs you to accept Google's model terms.
Why it's worth watching: people who dictate into Mac text fields and want the meeting, the screen, and the next sentence to stay on the machine. Version 0.8.6 is the build that adds Time Turner, Vidshots, and Cotypist together.
Boundaries: MIT License, copyright 2026 Pranav Hari. Free of charge under that license. Requires an Apple Silicon Mac running macOS 14.2 or later. Some models need a newer macOS. Signed with a Developer ID and notarized by Apple.
Discover Muesli on OpenNavo: opennavo.com/apps/muesli
#MacApps #Muesli #Dictation #MeetingNotes
muesli retweeted
I've dictated 3,000 words with muesli. It's fast, open source, on-device, and free to use. Try it: muesli.works
muesli retweeted
So I used Muesli today for recording and transcribing a 1.5 hour meeting. And it worked flawlessly. I've been using it for the past 3.5-4 weeks now. And every time, right, it's been amazing for dictation and it's, it doesn't miss anything? I use a local model, Whisper model
Replying to @progrium
muesli retweeted
wowza 51 contributors strong @fastspeech2text
muesli retweeted
free, fast, open source, on device and does a better job than their closed source alternatives
muesli.works
muesli retweeted
Sometimes i wonder why Muesli @fastspeech2text doesn’t get enough attention for practically the same thing everyone raves about.
This quoted post is unavailable.
muesli retweeted
I've dictated 110,000 words with Muesli. It's fast, open source, on-device, and free to use. Try it: muesli.works
muesli retweeted
Just use Muesli (hi @pHequals7), open source and healthier than granola (don’t quote me on that)
Got an email from @WisprFlow with step-by-step instructions to disable voice typing in @meetgranola.
I didn't even know Granola had that feature.
I turned it on and cancelled my WisprFlow subscription.
muesli retweeted
everything everywhere all at once
muesli retweeted
A few worth exploring:
• Muesli by @fastspeech2text: local-first meeting transcription and dictation
• FluidAudio by Levenza: a Swift SDK for local speech and audio AI
• React Native ExecuTorch by @swmansion: on-device inference for React Native apps
muesli retweeted
I've dictated 21,000 words with Muesli. It's fast, open source, on-device, and free to use. Try it: muesli.works
muesli retweeted
literally with muesli dot works exists
I'm looking to migrate from Granola and export away all of my data, which is being held hostage.
If you have a solution for this, please comment! Granola has this really shady strat where they give you free Granola for a like a year, then you find yourself not able to open your previous chats.
I don't want to pay Granola for the rest of my life. Please, help!
muesli retweeted
I was looking for a local-first alternative to whisper and granola at the same time, testing muesli.works/ right now it works like a charm! Very polished app, the onboarding was really smooth (not common for macos app 😂)
muesli retweeted
WisprFlow 一年 180 刀,Granola 还要另订,Muesli(1.1k★)一个应用把听写和会议转录都做了,Swift 原生 MIT 开源。语音转文字走 ANE 上的 Parakeet TDT,实测延迟 0.13 秒,录音不出本机。踩坑:实时转录默认关着,模型下载完不自动生效,得去设置里手动选。点击推文查看项目链接
muesli retweeted
Replying to @RichardDawkins
@RichardDawkins allow me to introduce you to @fastspeech2text


























