2.5%
Word error rate · lowest we tested
AI voice dictation · Windows & macOSYour intent,
Your intent,
perfectly on the page.
Hold a key, speak, and polished text lands in whatever app you're typing in, at a 2.5% word error rate.
ListeningYou said
um so I was thinking like maybe we could uh ship the the update you know on friday scratch thatlet's do monday
vbtm wrote
I was thinking we could ship the update on Monday, if that works.
~1s
From speech to polished text
3×
Faster than ElevenLabs Scribe v2
47
Languages supported
What makes it different
Dictation that disappears into your workflow.
Sub-second speed
Built in Rust with a streamlined AI pipeline. Words appear in around 1 second at just 2.5% word error rate, the lowest of any tool we tested.
Knows what you meant
It captures intent, not just words. Filler disappears, grammar tightens, and if you change your mind mid-sentence just say “scratch that.”
Context aware
Detects your active app. Formats emails in Outlook, prose in Word, and messages in Slack, with proper greetings, sign-offs, and structure.
Private by default
Your data is yours. We don't train on your voice. Audio is processed in transit and not retained after polish.
Why verbatim
I built this because Wispr Flow was the best thing I'd added to my workflow in years, and I wondered whether I could build something just as good.
A few months and one obsessive STT benchmark later, this is what I shipped. We picked Mistral Voxtral Mini, a transcription model that outperforms GPT-4o mini, Gemini 2.5 Flash, AssemblyAI Universal, and Deepgram Nova on accuracy, and runs roughly 3×faster than ElevenLabs' Scribe v2. Then we wired it to a polish step that knows which app you're dictating into, and lets you rewrite the AI's instructions yourself if our defaults aren't your style.
Why is it this cheap?
Verbatim is priced almost exactly where it stops making money. Each dictation has a real API cost (speech-to-text plus a small polish call), and Core is set just above that, with enough margin to keep the lights on and one developer fed. The honest reason: I don't think anyone else would build this at this price. The unit economics only work if you're a tiny team with no sales reps, VCs, or marketing agency. If we ever raise prices, it'll be because API costs went up, not because we got bigger.
The vbtm developerOne-person team · United Kingdom · replies to every email
Try it on your next email.
Free forever tier, no card required. 7 days of Pro on signup.