Spend Control for Deepgram
Deepgram transcribes at scale — per minute. Meeting and voice agents burn minutes fast. The guard caps the line.
Why Deepgram agents spend
Per-minute billing for transcription.
Meeting agents transcribe every session.
How it works
Guard call before transcription runs: APPROVED, BLOCKED, or FLAGGED.
Rules that fit
Per-team daily ceiling on transcription.
Velocity limit on retried sessions.
Guard call
import requests
r = requests.post("https://sipi.bot/v1/transactions/evaluate",
json={"amount": 6.0, "merchant": "deepgram", "category": "transcription"},
headers={"Authorization": "Bearer KEY"})
decision = r.json()["decision"]Minutes are a bill; the gate makes it a budget.
FAQ
Does this slow transcription?
No — ~5 ms per check.
Can I cap per team?
Yes — per-agent rules per team.
Related
Stop the next $12,400 night.
One API call (or MCP tool) in front of every agent transaction — APPROVED, BLOCKED, or FLAGGED, deterministic, ~5 ms, fully logged.
See plans — from $99/mo Try a live check