capability
verified July 2026
Usage accounting that is not a bill.
GET /v1/usage returns synthesized traffic per API key. On a metered vendor that endpoint is where your invoice comes from; here the invoice is lines times rate, so the usage API answers a different question — not “what do we owe” but “which workload talked.”
01
What per-key accounting is for
- A key per workload: the reminder job, the front desk, and the dialer each accounted separately.
- A key per environment: staging traffic never muddies the production picture.
- A key per client, for agencies and platforms — each portfolio’s traffic auditable on its own key.
curl https://tts.gandr.ai/v1/usage \
-H "x-api-key: gnd_reminders_job"
{
"key_id": "gnd_reminders_job",
"month": "2026-07",
"chars": 3211840,
"audio_s": 117780,
"requests": 41210
}
# one key per workload — the reminder job's month,
# readable on its own, reconciled against nothing
02
What it deliberately is not
A meter. Usage here never reconciles against an invoice, because characters and minutes are unmetered on a line — the figures are operational truth, not billing evidence. Nobody opens this endpoint at month-end holding their breath.
03
Where keys come from
Keys are issued on the demo call and authenticate with either header — x-api-key: gnd_… or Authorization: Bearer gnd_…. Usage is readable per key from the first request, so the audit trail starts when the traffic does.
04
Notes — an engineer's checklist
01Can I get usage broken down by voice or language?
The published surface is per key — which is why the working pattern is one key per workload, environment, or client. Cut the keys the way you want the report cut.
02Does heavy usage on one key affect my bill?
No. The invoice is lines times rate regardless of what the keys did. Usage figures exist so you can audit workloads, not so anyone can bill them.
03How fresh is the data?
Usage is readable per key from the first request onward — point a dashboard at GET /v1/usage and the reminder job, the desk, and the dialer each graph their own month.
See also
Related sheets.
capability
3
A text-to-speech API built for live calls
REST, SSE, and WebSocket TTS from one endpoint: 107 ms measured first audio, instant cloning from ten seconds, flat per-line pricing with unmetered characters.
use case
500
TTS for debt-collection voice agents
Collections dialers speak required disclosures on every connected call. The cost arithmetic of scripted outbound, per-key audit trails, and provenance watermarks.
capability
never
Provenance and privacy, applied at generation
The policy surface of the engine: an inaudible watermark on every sample, reference audio that never trains, and per-key audit trails at GET /v1/usage.
Your script, this API, thirty minutes — and the recording leaves with you.
Hear it on your own script