who it’s for
verified July 2026
Gandr for educators and course builders.
Teaching material is wrong on a schedule nobody controls. A regulation moves, a tool changes its interface, a figure is a year out of date. What decides whether the audio keeps up is how much a correction costs.
01
Why update cost sets curriculum quality
When re-voicing a module has a price, the sensible move is to batch fixes and ship them annually, and the audio drifts out of date between releases. When it has no marginal price, the fix ships the week the content changed.
The same arithmetic decides localisation. Under a meter every extra language multiplies the whole character count again, so the smaller cohorts get text while the large ones get audio.
02
The controls a lesson needs
- Spell tags for formulas, codes and acronyms — read character by character rather than mashed into a word.
- A pronunciation dictionary for domain terms, fixed per request rather than after a student complains.
- Break tags to hold a beat where the slide changes, so narration and deck stay in step.
- One cloned instructor voice across every localised edition.
03
Notes — an engineer's checklist
01Can one instructor voice cover every language?
Yes. One reference clip carries the identity across 23 languages, so localised editions keep the same teacher. Eight languages have published, dated specimens at /languages/.
02How do we handle formulas and acronyms?
Wrap them in a spell tag and they are read character by character with letter and digit groups paced naturally; a pronunciation dictionary handles terms that should sound a specific way.
See also — related sheets
use case
$0
The course that can afford to be updated
Course audio on a flat seat: when re-narrating a module costs nothing per character, curriculum updates stop being an annual event.
use case
2,000
A chapter costs the same as a sentence
Long-form narration on an unmetered seat: a whole book is a loop over requests, and the length of the book stops being a line item on the invoice.
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.
use case
24 kHz
A reading voice that is not a punishment
Accessible audio on an unmetered seat: read the whole document, every time, without the character count arguing for a shorter description.
A key and one seat to size it on, before anyone talks about a fleet.
Request access for your stack