who it’s for
verified July 2026
Gandr for creators.
A channel is a schedule. Something ships every week whether the script was finished on Monday or Thursday night, and the last hour before publishing is where voiceover either helps or gets in the way. A meter turns each re-render into a small decision; a seat removes the decision.
01
What the schedule actually costs
The bill for one video is never the problem. The bill for a weekly channel that re-cuts each script three or four times is the same number multiplied by the thing you cannot reduce, arriving in the month you were trying to raise the quality.
On a seat the fourth pass costs what the first did. That changes which edits you are willing to make at 11pm — the alternate open, the tighter mid-roll, the line you would have left alone.
02
What you get on one seat
- Your own cloned read from ten seconds of reference — no enrolment, no per-voice fee.
- Long scripts as a loop of requests, with characters unmetered, so a twenty-minute video costs what a two-minute one costs.
- Break and spell tags to place a beat on a cut or read a product code cleanly.
- A seed, so re-rendering one fixed sentence leaves the rest of the track untouched.
03
Notes — an engineer's checklist
01Can I use my own voice?
Yes — five to ten seconds of you speaking becomes a request parameter. Cloning requires a consenting speaker, and every clip carries an inaudible provenance watermark applied at generation.
02Is one seat enough for a channel?
For one person editing one video at a time, yes: a seat runs one stream of speech at a time and everything on it is unmetered. A second editor working in parallel is a second seat.
03What if I publish in more than one language?
One cloned reference carries the same identity across 23 languages, so the channel keeps its voice in every market rather than handing it to a different narrator.
See also — related sheets
use case
$0
A narrator who never gets tired of take nine
Script-to-voiceover on a flat seat: re-render the whole video after an edit, in your own cloned voice, without a character count deciding how long the script runs.
use case
107 ms
A show that can be re-cut after it is recorded
Podcast audio on an unmetered seat: intros, ad reads, corrections and localised editions re-render as often as the show needs, at no cost per character.
capability
0
Voice cloning from ten seconds, in the request
Zero-shot voice cloning: a ten-second reference rides inside each request, no training job, and the identity holds across 23 languages.
capability
7.9
Delivery is a request parameter
The prosody dial moves measured pitch range from 10.1 to 18.0 semitones on one voice, one sentence — continuous per-request control on the production API.
A key and one seat to size it on, before anyone talks about a fleet.
Request access for your stack