Health broadcast
Ku zo asibiti gobe daga karfe takwas na safe zuwa karfe biyu na rana, domin allurar rigakafi ga yara 'yan kasa da shekaru biyar. Ba a biyan komai.
Kanya
Speech synthesis
Namu-Voice
Turn Hausa text into speech people want to listen to.
Features
Namu-Voice reads Hausa the way people say it, not the way it is written. Tone, vowel length and the rhythm of a sentence all come through. And every voice it uses is one somebody agreed to give.
Realistic expression
Organic pacing, tone and emotional range that sound like a person, not a text-to-speech engine.
Voice
Kanya
Joy
Emotion
Instant voice matching
Give it a short clip of someone speaking, with their permission, and it can read in their voice. No training run, no dataset to gather.
Built for the people
Stable, high-fidelity output that holds the same speaker across an hour of audio.
Natural across eight Hausa dialects
Hausa is not one accent. The model is trained and measured on the varieties people speak across Niger, from Maradi to Zinder.
MaradiZinderTahouaTessaouaKonniDogondoutchiMadaouaMirriah
Using the model
Hausa speech, live and on demand.
Samples generated with Namu-Voice
Market report
Farashin gero a Dawanau ya kai dubu talatin da biyar kan buhu. A Jibia kuwa, dubu talatin da uku. Masara ta tsaya kan dubu ashirin da takwas.
Marke
Story reading
Da daddare, iska ta fara busawa daga gabas. Kowa ya shiga gida, sai tsohuwa daya tak ta zauna a bakin kofa tana jiran dawowar jikanta.
Baobab
Performance
Leading on naturalness in Hausa
Namu-Voice holds its speaker across long passages and starts speaking fast enough to hold a conversation.
Listen across dialects
Dig deeper
Try Namu-Voice
Namu Playground
Type a line of Hausa and hear it back in any of the four voices.
Namu API
Stream speech into your own product, with the same voices and consent controls.
All Namu models
Interpretation, transcription and the end-to-end voice agent.