← All tutorialsAudio

Voice-note transcriber

User drops a 5-minute m4a in Slack. Transcribe with Deepgram, summarize with GPT-4o-mini, post back as a thread reply.

Cost~$0.04 per 5-min clip
Operations
speech.transcribetext_generation.generatechat.thread_reply

Prerequisites

  • speech.transcribe enabled (Deepgram-backed).
  • Slack bot wired with chat.* permissions.

Walkthrough

1. Transcribe

`audio_url` must be publicly fetchable (or a presigned S3 URL).

bash# Deepgram returns text + word timings.
curl -X POST https://www.upivia.com/v1/service-requests \
  -H "Authorization: Bearer $AGENT_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "service":"speech",
    "operation":"transcribe",
    "payload":{"audio_url":"https://cdn.example.com/note.m4a","language":"en"}
  }'

2. Summarize

Pipe the transcript into GPT-4o-mini, ask for 3 bullets + action items.

3. Reply in thread

Use chat.thread_reply with the original message's `thread_ts` so the summary lands beneath the voice note.

Next steps

Audit every call at /audit-logs, watch spend at /usage, and tune budgets per service on the agent's page.

Create an account →