glossary
verified July 2026
TTS pricing: per character vs flat rate
Metered pricing bills what your agents say — per character or per minute. Flat pricing bills the capacity to say it — per concurrent line. The break-even between them is arithmetic, not philosophy.
01
The two models, defined
Published character meters run $30 to $100 per million characters (July 2026), and a spoken minute is about 1,000 characters — so a metered agent costs 3 to 10 cents per spoken minute, forever. A flat line is $150 a month for one concurrent conversation, characters unmetered. The full four-workload arithmetic, with the crossing drawn, lives in the lab report; this page is about what each model does to the product built on it.
02
What the meter does to a product
- It prices growth: close a big customer and the voice bill doubles with the win.
- It prices quality: when characters cost money, someone shortens the agent’s answers.
- It prices interruptions: every barge-in bills characters nobody heard.
- It breaks forecasts: “depends how much people talk” is not a line item.
03
What flat pricing costs you
Idle capacity. A line you bought and did not talk on is money spent — which is why low-volume, single-workload traffic belongs on a meter, and why our own reminder-workload page says so. Flat wins when lines are busy or shared across workloads; meters win when they are neither.
See also
Related sheets.
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.
glossary
1 line
Concurrent lines
A line is one simultaneous conversation. Why concurrency — not minutes — is the real capacity unit for voice agents, and how metered plans cap it quietly.
use case
2,000 min
TTS for appointment reminders and notifications
Reminder traffic is bursty and short-form. Where a meter genuinely wins, this page says so — and where burst lines beat both, it shows the arithmetic.
Every term on this page is measurable on a live call — watch the readout while your own script synthesizes.
See the measured figures live