The language

Azerbaijani as the design, not as a setting

Chinar is not a multilingual model with Azerbaijani bolted on, and not Turkish with an accent. The language is what it was built around — which is the difference between a transcript you can act on and one that merely looks finished.

Overview

Azerbaijani speech recognition

Most speech engines treat Azerbaijani as a neighbouring case of a language they already handle, and the output reads fluently enough that nobody checks it. Chinar is built for Azerbaijani itself, in two sizes that share the same command of the language: Chinar-L for transcripts people read, Chinar-F for transcripts machines read. Both run on your own infrastructure.

What it does

What building for the language changes

Azerbaijani is the design target, not a language code passed to a general model.

Two sizes share the same language expertise, so their transcripts agree.

Chinar-L scores 87% word accuracy on clear Azerbaijani speech, measured against reference transcripts from a commercial speech system.

Trained on genuine call-centre recordings rather than read speech.

Runs on your own infrastructure, with no audio sent to a third party.

How it works

How a recording becomes text

1Choose the size: Chinar-L for readable transcripts, Chinar-F for volume
2Deploy on your own infrastructure, inside your own network
3Send audio through the API — call recordings, interviews, meetings
4Read the Azerbaijani text, or feed it into search, analytics or reporting
FAQ

Common questions

Is this a fine-tuned version of a general speech model?

Chinar is built for Azerbaijani from the ground up rather than adapted from a related language.

How accurate is it?

Chinar-L reaches 87% word accuracy on clear Azerbaijani speech, measured against reference transcripts from a commercial speech system. Noisier audio scores lower.

Which size should we start with?

Chinar-L when a person reads the transcript; Chinar-F when the volume matters more than the punctuation.

Explore more

More of what Chinar does

Get started

Hear Chinar on your own recordings

Request a demo and we will run Chinar against your own Azerbaijani audio — Chinar-L for the transcript people read, Chinar-F for the volume.