Vocard
Voice recordings turned into organised notes — transcript, summary, action items, and translation.
- React
- OpenAI Whisper
- LLM post-processing
- Subscription billing
Why it needed building
Anyone who records meetings, lectures, or interviews ends up with audio nobody ever listens to again. Transcription alone does not fix it: a wall of text is not notes.
What we built
- In-browser recording and file upload across MP3, WAV, M4A, OGG, WebM and AAC
- Speech-to-text on OpenAI Whisper, tuned for accented English and ten more languages
- AI post-processing into summary, action-item, and translation cards
- SuperSummary — one summary generated across two to ten separate notes
- PDF and Markdown export, and re-processing a note with newly appended audio
- Free and Pro plans with usage metering and billing
What it looks like
One tap to record
Pick a language and start talking, or upload audio you already have.
Every recording, filed
Each recording becomes a titled note, searchable and sortable.
Free to Pro
Metered free tier with an upgrade path, billing included.
Settings that stick
Plan, notifications, theme, and language in one place.
This is what our Startup MVP work looks like
A working product, scoped tight enough to launch and built well enough to keep after it works.
See Startup MVPTell us what you are building
A free 30 minutes. If we are not the right people for it, we will say so and point you at someone who is.
No sales sequence, no mailing list. We reply within 24 hours.