English is my second language, and while speaking it is fine, writing and polishing takes me noticeably longer than native speakers. I was a Wispr Flow user for a while - liked the idea, but it had too many missing pieces and some things that genuinely annoyed me.
So I built my own. It's called Voicr - a native Mac app. Small, fast, and the whole thing is 1.2 MB. It fits on a floppy disk if you still remember what that is. 😄
Here's what it does:
- Dictation. Hold FN, speak, release - text is transcribed and pasted into whatever field you're in.
- Notes. Hold FN + Control and it saves a note directly, no polishing, just the raw transcription.
- Smart rules. Set a custom prompt per app - casual for Slack, formal for email - and it rewrites your dictation automatically when you switch context.
- Text correction. Select any text, press ⌥Space, pick a correction rule, and it rewrites in place. No copy-pasting to ChatGPT and back.
- Translation. Speak in any language, get the output in another. I use this more than I expected.
Why it is cool?
- Mac native. Small, fast and efficient. F! electron :-)
- Privacy first. It doesn't store any data, everything is stored locally.
- It doesn't use OpenAI or any other providers. Hosted models in cloude: Whisper and Llama.
- You can create as many corrections and rules as you want. For different apps for different proposes.
So it's a dictation app, a note taker, a writing assistant, and a translator — in 1.2 MB.
What do you think? I really love what I've made and use it everyday.
Link: voicr.pro
This is really clean — FN-to-speak + app-specific rewriting is a great UX combo.
Love the “no copy-paste to ChatGPT” angle — that’s a real daily friction point. And 1.2MB native is impressive.
I think showing real before/after writing speed gains could be a strong hook.
Also, if you’re building tools like this — $19 puts it in real competition. Tokyo trip + $500 min guaranteed.
Round just opened: tokyolore.com 🚀
The product sounds useful, but I think the pricing question is slightly downstream from the positioning question. "Saves 30 min/day" is logical, but the stronger pitch is probably about reducing friction in the exact moments where writing feels slow or context-switching breaks flow. If people try it and immediately feel "I don't have to open three other tools anymore," then $3/mo is easy — if they have to think too hard about what it replaces, conversion gets harder.
Yeah, I am trying many different pitches right now, one of them speed of talking instead of typing.