Skip to content
VoxNotch
Join Beta
Menu

API

A private audio transcription API for workflows you operate.

Developers do not just need a transcript endpoint. They need a queue, ownership boundary, webhook destination, export shape, job history, and a way to prove what ran when a private recording moved through automation.

Primary keyword
Audio transcription API
Inputs
Upload, browser, watch folder, API
Outputs
Transcript, captions, reports, webhook events
Operation
Self-hosted jobs and run history

API transcription workflow with watch folder, job queue, webhook, and transcript export.

Shape

workflow
upload or watch folder -> queued STT job -> transcript -> export -> webhook destination

The valuable API is the workflow state around transcription, not just a synchronous text response.

Curl shape

shell
export VOXNOTCH_URL="https://voxnotch.example.test"
export VOXNOTCH_API_KEY="vtk_example_do_not_use_this_key"

upload_response="$(
  curl -sS -X POST "$VOXNOTCH_URL/recordings/upload" \
    -H "X-API-Key: $VOXNOTCH_API_KEY" \
    -F "file=@./meeting.wav" \
    -F "coverage=balanced"
)"

job_id="$(printf '%s' "$upload_response" | jq -r '.job_id // empty')"
[ -n "$job_id" ] && curl -sS "$VOXNOTCH_URL/recordings/upload/$job_id" \
  -H "X-API-Key: $VOXNOTCH_API_KEY"

Fake host, fake key. The real API key is created from a signed-in browser session and the upload response can be a duplicate with no new job_id.

Build against the instance you control.

Join BetaDeveloper surface