Tambourine vs Built-in Dictation
Why AI formatting beats raw transcription.
Quick Summary
macOS and Windows built-in dictation transcribes your speech word-for-word, including every "um", "uh", and false start. Tambourine uses AI to clean up your speech: removing filler words, adding proper punctuation, applying your personal dictionary for technical terms, and producing clean, publication-ready text.
See the Difference
Built-in Dictation
"so um I think we should uh use the API to like fetch the data and then um parse it"
Tambourine
"I think we should use the API to fetch the data and then parse it."
Feature Comparison
| Feature | Tambourine | Built-in Dictation |
|---|---|---|
| Removes Filler Words | Yes (um, uh, like, etc.) | No |
| Auto Punctuation | AI-powered | Basic / voice commands |
| Personal Dictionary | Yes (technical terms, names) | Limited |
| Backtrack Corrections | "actually" / "scratch that" | No |
| Custom Formatting Rules | Editable prompts | No |
| List Formatting | Auto-detects lists | No |
| Choose STT Provider | 10+ options | System default |
| Works Offline | With Whisper + Ollama | Yes |
| Open Source | Yes (AGPL-3.0) | No |
Why Choose Tambourine
Clean Output
AI removes filler words, false starts, and verbal tics. Get publication-ready text without editing.
Personal Dictionary
Teach it technical terms, API names, and proper nouns specific to your work.
Customizable Rules
Edit the formatting prompts to match your writing style and preferences.
Open Source
Full source code access. Self-host for privacy, or contribute improvements.