Norynta Data › Language data index › Hakka
Hakka speech data for AI
How much open Hakka speech data exists, where the gaps are, and how to get consented Hakka data collected.
There is no validated open Hakka speech in Common Voice yet. Hakka is one of 163 Common Voice languages with zero validated hours, so almost any Hakka speech data has to be collected new.
Source: Mozilla Common Voice language statistics, fetched 2026-11-05. Figures cover the open Common Voice corpus only.
Need Hakka data?
We recruit native Hakka 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 Hakka?
Hakka
Join the contributor waitlist and we'll email you paid Hakka 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 Hakka speech data for AI training?
Norynta collects Hakka 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 0 validated hours of Hakka (as of 2026-11-05).
How much open Hakka speech data is available?
As of 2026-11-05, Mozilla Common Voice lists 0 validated hours and 0 recorded hours of Hakka from 0 speakers, ranking 322 of 445 languages by validated hours.
Can I get paid to record Hakka?
Yes. Native Hakka speakers can join Norynta's contributor waitlist at collect.norynta.com/join. We email paid recording and writing projects in Hakka when they open, and pay once the work is accepted.
Languages with similar coverage
- Gorani 0 h
- Gronings 0 h
- Gujarati 0 h
- Haitian 0 h
- Heng Hua 0 h
- Highland Puebla Nahuatl 0 h
- Hiligaynon 0 h
- Hunsrik 0 h