who it’s for
verified July 2026
Gandr for accessibility teams.
Accessibility is where a meter does its quietest damage. Every affordance is extra characters — the longer image description, the confirmation read back, the error explained rather than announced — and each one is a small saving that adds up to a product someone cannot use.
01
Where the meter edits the experience
The pressure is always toward saying less, and it never arrives as a decision anybody would defend out loud. It arrives as a prompt that says be concise, applied to the text a screen-reader user depends on.
On a seat the extra sentence costs nothing, so the question returns to what a person actually needs to hear.
02
What matters technically
- Spell tags read codes and reference numbers character by character — the difference between a usable read-back and a guess.
- A low temperature holds a steady, predictable delivery, which is what a voice someone listens to all day should do.
- Break tags let structure be heard: a real beat between a heading and its content.
- 24 kHz mono by default, so the audio is full-rate rather than the compressed telephony band.
03
Notes — an engineer's checklist
01Is this a screen reader?
No — it is the synthesis layer a screen reader or an application calls. Navigation, semantics and focus handling stay in your product or in the assistive technology someone already uses.
02What sample rate does audio come back at?
24 kHz mono PCM by default. Set output_sample_rate if your playback path wants something else.
See also — related sheets
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.
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
$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.
capability
3
A text-to-speech API built for live audio
REST, SSE, and WebSocket TTS from one endpoint: 107 ms measured first audio, instant cloning from ten seconds, and a flat per-seat price with usage unmetered.
A key and one seat to size it on, before anyone talks about a fleet.
Request access for your stack