capability
verified July 2026
The pilot: a month on real traffic.
Nothing on this site asks you to believe anything — every figure reruns on demand. The pilot is the same posture at contract scale: route one real queue through 20 lines for a month, and decide on your own numbers.
01
The terms
Fig. — The whole agreement
20 lines
unlimited minutes, instant cloning, all three streaming surfaces
30 days
no contract — turn it off at the end and keep the data
$1,000
the first month, credited toward your first annual invoice
A direct line to the founders runs for the whole integration — the pilot is also how we learn your workload.
02
Why 20 lines
Twenty lines is enough concurrency to shadow a real queue — a support pod, one IVR segment, a regional desk — rather than a toy. The month answers the questions this site can only argue: your busy-hour line count, your burst days, your p95 under your own prompts. If the numbers do not hold, the honest outcome is that you leave with a measured no.
03
What to bring, what you leave with
- Bring production prompts — the pilot runs your script, not our demo copy.
- Bring your current TTS invoice, so the month ends in a comparison, not an impression.
- Bring ten seconds of the voice you want cloned, and hear it on your own traffic.
- You leave with the recording, the per-key usage record of the month, and a one-page plan — worst case, leverage on your current vendor.
04
Notes — an engineer's checklist
01What happens at the end of the 30 days?
Either the numbers held and the $1,000 credits toward your first annual invoice, or they did not and you turn it off — keeping the recordings, the usage data, and the comparison.
02Can the pilot run beside my current vendor?
That is the recommended shape: shadow one queue while the rest of your traffic stays put. The request shape makes the split a routing rule, not a migration.
03Is 20 lines negotiable?
The pilot is deliberately one size — big enough to be real, small enough to decide fast. Fleets that need a different shape start with the same 30-minute call.
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.
lab report
5,000 min
Per-character pricing, run against flat lines
Four voice workloads run through a $30–$100 per-million-character meter and through flat lines, with the break-even stated and the cases where the meter wins.
use case
43,200 min
TTS for AI receptionists
A 24/7 receptionist line speaks up to 43,200 minutes a month: $1,296 to $4,320 on a character meter, $150 flat. The arithmetic for always-on voice.
capability
per key
Usage accounting that is not a bill
GET /v1/usage returns synthesized traffic per API key. Why usage data here is operational accounting rather than the source of your monthly invoice.
Your script, this API, thirty minutes — and the recording leaves with you.
Hear it on your own script