Verbatim
AI voice dictation for Windows and macOS — indie-built, half
It's taken about 4 months, and today I've shipped verbatim - vbtm.app. An AI voice dictation app for Mac and Windows.
I came across Wispr Flow and thought it was great, but expensive. I thought I'd try build something similar myself. There isn't a custom model that uses your data to train, instead we use Mistral Voxtral Mini - the highest scoring and most accurate transciption service available.
I've set pricing as lean as I can (hopefully just above API costs). The unit economics work because it's just me, no fancy setup, and I built this for me. May as well share the love!
Claude's telling me I should be spouting off about all my proudest moments and biggest errors when building this. If you actually care, happy to detail, but getting this live did invovle a fair amount of screaming into the ether as well as quite a lot of very satisfying moments when the thing actually did what it was supposed to!
I think it's great, let me know what you think. Free 60 day trial if you want to give it a go - msg me (I think you can do that), but just please let me know if anything doesn't work as expected!
1 Comment
This is a solid launch, but I think the positioning can be sharper than “Wispr Flow but cheaper.”
Cheaper gets attention, but it can also make people judge the product mainly on price. The stronger angle is probably something like:
“fast voice dictation for people who write all day but do not want another expensive AI subscription.”
That gives you a clearer first user: founders, writers, consultants, devs, and operators who constantly turn thoughts into text.
The 60-day free trial is generous, but I’d be careful that it does not delay the real signal too much. The useful test is not whether people try it. It is whether they use it repeatedly for real work after the first week.
I’d probably track one thing very closely: how many users dictate something useful three separate times in the first 7 days.
Happy to put a tighter version in writing if useful. The main thing I’d map is the Wispr comparison, first user segment, pricing message, and trial-to-paid path.