Bootstrapping speech for low-resource markets: public corpora, real calls

The cold-start problem

Every new language a speech model learns starts cold. For English there are hundreds of thousands of hours of transcribed audio in the open; for Nigerian Pidgin, Hausa, Yoruba or Igbo there is a small fraction of that, and almost none of it drawn from real phone calls. A model trained from scratch on what little exists is fragile, and collecting enough of your own labelled audio to replace it takes months. The practical question is not whether to use public data, but how to use it without inheriting its blind spots.

What the public corpora actually offer

A handful of datasets are worth knowing for West African speech:

• NaijaVoices — roughly 1,800 hours across Hausa, Yoruba and Igbo, from thousands of speakers. The largest local-language resource by far.
• Mozilla Common Voice — community-recorded speech including Nigerian Pidgin, released into the public domain.
• FLEURS — a smaller multilingual read-speech set that covers Hausa, Yoruba and Igbo with clean, consistent recordings.
• AfriSpeech / AfriVox — pan-African accented English, useful for the Nigerian-English track and as an evaluation benchmark.

Together these are enough to give a model a first, rough sense of each language's sounds and spelling — a warm start instead of a cold one.

The two catches nobody mentions

The first is licensing. The largest set, NaijaVoices, is released under a non-commercial licence. That makes it excellent for research and for measuring how good a model is, but it cannot simply be poured into a model that ships inside a commercial product. For anything customer-facing we lean on the permissively licensed sets — Common Voice, FLEURS, AfriSpeech — and treat the non-commercial data as a benchmark, not a training set, unless a commercial licence is agreed.

The second is the kind of speech. Almost all public audio is read speech: someone reciting a prompt in a quiet room into a decent microphone. Real business calls are the opposite — narrowband 8 kHz audio over the phone network, compressed by the carrier, full of background noise, interruptions, and speakers switching between languages mid-sentence. A model that has only heard read speech is confident and wrong the first time it meets a real call.

How we use them

So we treat public corpora as a bootstrap, not a destination. The license-clean sets pretrain the base model — they teach it the shape of each language cheaply, before it has heard a single real call. Then the model is fine-tuned on real, consented telephony collected through our own pipeline, which is where the accuracy customers actually feel comes from. Public data does one valuable thing: it shrinks how much of our own labelled audio we need before the model is useful. It does not remove the need for that audio.

Why the real calls still matter most

The reason is simple: a model is only reliable on the kind of audio it was trained on. The data the product runs on — real conversations, on real networks, in the languages customers actually speak — is the data worth owning. That is why our collection is built around consent and de-identification from the first step rather than scraped from the open web, and why the verified set is reviewed by people who speak the language. Public corpora get a model to the starting line. Real, consented calls are what let it finish the race.

Where this goes

The near-term work is specific: pretrain each language track on the permissively licensed public data, stand up honest evaluation against held-out real calls, and grow the verified telephony set language by language. The measure of success is not how much public data we can absorb, but how little of our own we need before a model understands a real Nigerian call.