Buying AI training data from Nigeria? Email [email protected]Request a free sample

Hausa (ha) · Lagos, recording to order

Hausa speech data for ASR and TTS, from consenting speakers.

Read and conversational Hausa from ID-checked speakers who consent on record, transcribed in standard Latin (Boko) spelling with the hooked letters, and region recorded for every clip.

Email [email protected] Request a dataset

Free 10-hour sample first. A person replies within one working day.

A Nigerian woman speaking into her phone at home

Where it is spoken

Hausa at a glance.

Spoken across northern Nigeria (Kano, Kaduna, Katsina, Sokoto and beyond) and in Niger, Ghana and other West African countries, as a first language and as a widely used trade language.

Code: ha (ISO 639-1)
Script: Latin (Boko) with hooked letters ɓ, ɗ, ƙ and ʼy; Arabic-based Ajami exists and can be discussed
Tone and length: Hausa marks tone and vowel length in teaching texts but rarely in everyday writing; we add them on request
Dialects: Standard (Kano-based) Hausa by default; other regional varieties tagged when you need them

Why it is hard for models

What speech models get wrong.

Hooked letters

ɓ, ɗ and ƙ are often typed as b, d and k online, so clean transcripts with the right letters are scarce.

Tone and vowel length

Both carry meaning but are rarely written, which makes text-to-speech and pronunciation hard.

Arabic and English loanwords

Everyday Hausa borrows from Arabic and English; transcription guides must handle both.

ASR and TTS

Read prompts or conversation? Usually both.

For speech recognition (ASR)

For ASR we record many speakers on read prompts and conversation, transcribed with the hooked letters so your text normaliser has clean targets.

For text-to-speech (TTS)

For TTS we record a few speakers for longer from a balanced prompt list, with tone and length marked on request, and voice-synthesis consent named explicitly.

Speaker diversity

A fair mix of real speakers.

Speakers by region of origin, age band and gender to a quota you set
Lagos-based Hausa speakers first; northern cities such as Kano are on our next list
Every speaker ID-checked so one voice is one person
Market, transport, farming, family and service topics

Recording spec

How each clip is recorded and delivered.

Audio: WAV 48 kHz, 16-bit, mono per speaker (one track each in two-person sessions)
Tasks: Read prompts (your list or ours), natural two-person conversation, scripted service dialogues
Transcripts: UTF-8, one per clip, speaker turns and non-speech events marked; second-pass checks on every batch
Metadata: Language, age band, gender, region, device, consent ID and consent version per clip
Consent: Recorded in the speaker’s own language; voice synthesis (TTS) named explicitly when it applies
Delivery: Weekly batches over signed links, with a manifest and a QA report

Open data first

Honest advice: what to try before buying.

NaijaVoices (open, non-commercial)

NaijaVoices is an open speech dataset of Igbo, Hausa and Yorùbá released under CC BY-NC-SA 4.0, which does not allow commercial use. Good for research and benchmarks; licensed data is for products.

Source: NaijaVoices dataset card, Hugging Face

Open sets exist; check their licences

Hausa appears in open collections such as Mozilla Common Voice and academic corpora. Use them to start; we supply licensed, consented data to your spec.

Source: Mozilla Common Voice

Questions

Buyers usually ask.

Can you transcribe in Ajami (Arabic script)?

Our default is Latin (Boko) spelling with hooked letters. Ajami transcription can be scoped per order if you need it.

Are speakers recorded in northern Nigeria?

Today our speakers are in Lagos, which has a large Hausa-speaking community. Kano is on our list of next cities, opened when buyer orders need it.

How do I order Hausa speech data?

Email [email protected] or use the request form with the hours, read/conversation split, speaker mix and transcript rules you need. We record a free sample of about 10 hours to your spec first. Pricing is quoted per approved hour after scoping.

How do speakers consent?

Each speaker passes an ID check and records consent in their own language before any paid task. If the data will be used for voice synthesis (TTS), that use is named explicitly. Speakers can withdraw from future deliveries.

Keep exploring

Tell us what you need. We send a free sample first.

Email [email protected] with the data type, hours, languages or trades, labels and format. Pricing is quoted per approved hour after scoping.

Email us Request form