Zumbo
Open source dictation for Mac with local AI, instant speed and complete privacy
Zumbo turns your speech into text on your Mac in about a fifth of a second using local AI models, with 21 vocabulary packs, notes, meeting transcription and
About this launch

Zumbo is an open source dictation app for macOS that uses local AI models running on your Mac to convert speech to text in about 200 milliseconds, with no server and complete privacy. It works in any app where you can place a cursor—from terminal and code editors to email, chat and documents. The app includes 21 vocabulary packs for fields like development, legal, medical and design, learns from your corrections to improve accuracy over time, and offers a Notes feature with reminders plus Meeting mode for transcribing conversations with speaker labels. All processing happens locally on your Mac; internet is only used once to activate your license and once daily to check for updates. Zumbo is a one-time purchase with free updates, available in 25 European languages, and runs on Apple silicon Macs with macOS 14 or later.
What it does
- Dictation into any app with one keyboard shortcut
- Local speech processing in about 200 milliseconds
- Automatic clean speech with filler word removal
- Learn corrections to improve accuracy per user
- 21 vocabulary packs for specialized fields
- Custom vocabulary packs from CSV or JSON files
- Notes with searchable history and time-based reminders
- Meeting mode with speaker labels and timestamps
- Supports 25 European languages
- Runs completely offline with no account required
Pricing
Pricing model: Paid
One Mac: $15 (or $9 for first 10 buyers); Three Macs: $20; Team (10 Macs): $79. 3-day free trial with everything unlocked. Free updates. 14-day refunds.
Shopper questions
Does it need the internet?
No. The engine runs on your Mac. Internet is used once to activate your license and once a day to check for updates.
How accurate is it?
On the company's own recordings, technical terms come out right 9 times out of 10 with the vocabulary packs enabled, and every correction you make improves accuracy for your voice. Normal sentences are where it performs best.
Is it a subscription?
No. You pay once and keep it. Updates are free.
Is it open source?
Yes. The whole app is on GitHub under GPL-3 at github.com/claudiusararu/zumbo. You can read every line, build it yourself, or buy the signed build with the speech model inside and updates through the app.
Which languages?
25 European languages, English first. Pick your main language in setup; Zumbo then writes only in that alphabet.
Zumbo has a free listing. The owner can unlock the full profile with a paid launch for $5:
- Founder story (locked)
- Milestones (locked)
- Demo video (locked)
- Dofollow website link (locked)
Shoppers also viewed
- TokScript
Audio, Voice & Podcasting
- Relayo
Audio, Voice & Podcasting
- hearlog
Audio, Voice & Podcasting
- Audionarrator
Audio, Voice & Podcasting