Can I code this? A Call Recording and Transcription Tool
Recording a call and running it through an existing speech-to-text API is genuinely achievable. Actually connecting to phone lines to capture the call in the first place needs real telephony infrastructure, not just transcription.
Borderline
Effort: a weekendRunning cost: $0/mo to start
Why CanICodeThis says Borderline
Once you have an audio recording, sending it to an existing speech-to-text API and displaying the transcript is a bounded integration
For calls made through an existing video/voice tool with a built-in recording export, this is genuinely a weekend project
Storing and searching past call transcripts is ordinary CRUD
The parts that'll cause problems
Actually capturing a live phone call (not a VoIP app call you already control) needs real telephony infrastructure — a service like Twilio exists specifically because this is a genuine integration surface, not something to build from scratch
Legal consent-to-record requirements vary by jurisdiction and carry real compliance weight
Real-time transcription during a live call, rather than after it ends, needs streaming audio infrastructure
How I'd build this
frontend
Next.js
database
Supabase
hosting
Vercel
MVP scope
Upload a call recording (exported from an existing calling tool)
Send to an existing speech-to-text API for transcription
Store and search past transcripts
Postpone to v2
Live telephony integration via a service like Twilio
Real-time transcription during the call
Automatic consent-to-record disclosure per jurisdiction
What it actually costs to build
AI agent — with a subscription (Claude Pro/Max, Cursor, etc.)
$0 marginal
AI agent — pay-per-use API, no subscription
$43–$86 one-time
Hosting, once it's running
$0/mo (free tier)
Domain name, if you want your own
~$12/yr
Existing tools solving a similar problem typically run $16.99–$16.99/mo — see how they compare below.
The build prompt
Build me a call transcription tool: upload a call recording exported from an existing calling app, send it to an existing speech-to-text API, and store the transcript, searchable by keyword. Don't build live telephony capture from scratch — use an existing telephony API like Twilio if you need to record calls directly, and make sure consent-to-record requirements for your jurisdiction are handled correctly before recording anyone.