Norynta Data › Language data index › Mingrelian
Mingrelian (მარგალური) speech data for AI
How much open Mingrelian speech data exists, where the gaps are, and how to get consented Mingrelian data collected.
That is enough to start experiments, but far short of production speech recognition, which typically uses thousands of hours. Conversational, accent-balanced and domain-specific Mingrelian audio remains scarce.
Source: Mozilla Common Voice language statistics, fetched 2026-11-05. Figures cover the open Common Voice corpus only.
Need Mingrelian data?
We recruit native Mingrelian speakers and collect to your spec: read prompts or natural two-person conversation, transcripts in the native script, speaker age, gender and region labels, noise and synthetic-voice checks, and a consent record for every file. Shared or exclusive licence.
Speak Mingrelian?
მარგალური
Join the contributor waitlist and we'll email you paid Mingrelian recording and writing projects when they open. You record on your phone and are paid once your work is accepted.
Questions
Where can I get Mingrelian speech data for AI training?
Norynta collects Mingrelian speech data to order from consenting native speakers: read prompts, natural conversations, transcripts and private evaluation sets, with a consent and origin record for every file. Request it at collect.norynta.com/buy. For open data, Mozilla Common Voice has 12 validated hours of Mingrelian (as of 2026-11-05).
How much open Mingrelian speech data is available?
As of 2026-11-05, Mozilla Common Voice lists 12 validated hours and 13 recorded hours of Mingrelian from 92 speakers, ranking 126 of 445 languages by validated hours.
Can I get paid to record Mingrelian?
Yes. Native Mingrelian speakers can join Norynta's contributor waitlist at collect.norynta.com/join. We email paid recording and writing projects in Mingrelian when they open, and pay once the work is accepted.