Skip to content

Build once. Reach everywhere.

One API for natural speech, voice cloning, speech to text and translation across 73 African languages in 44 countries.

Languages
73
Countries
44
Voice APIs
4
English (Nigeria)Nigerian English · John
STORM v1

Hello, I'm John. I read your text aloud in clear Nigerian English, and your app can start playing it in under a second.

Trusted by teams building for African markets

  • Storipod
  • EndowPay
  • University of LagosUniversity of Lagos
The platform

Four voice APIs. One key for all of them.

Speech is the product. Transcription and translation complete the loop, so you can listen, understand and answer in one integration.

A speaker turning text into a sound wave
POST /v1/speech

Text to speech

Natural voices for 73 African languages.

  • Hand-tested native voices for Yoruba, Hausa, Igbo, Pidgin and Nigerian English
  • Every other language on STORM v3
  • Streams audio as it is generated
Hear John, Nigerian English

Streaming

Playback starts while the rest of the sentence is still being generated.

"stream": true

Text normalisation

Prices, dates, times and percentages are read the way people say them.

₦5,000 → “five thousand naira”

Cloning from seconds

A short, clean clip is enough to create a voice you can reuse.

4–8 s clip → voice_id
In production

Hear it before you build it. Every clip is real STORM output.

Every clip on this page came straight out of the API. Press play on any of them.

One API for every message you send.

Alerts, menus, lessons and ads, in the language your customer grew up speaking.

STORM v1

Native voices, checked by native speakers.

Yoruba, Hausa, Igbo, Pidgin and Nigerian English voices, tested by people who speak them.

  • Nigerian EnglishJohn
  • YorubaMorenike
  • HausaAmina
  • IgboAdaeze
  • Nigerian PidginJohn
  • Naira, dates and numbers spoken in full
  • Streaming for a fast first sound
  • Hand-tested on STORM v1

Usage your finance team can read.

Limits, usage and invoices for every service, in one dashboard.

Text to speech412,880 / 600,000 characters
Speech to text38 / 120 minutes
September invoicePaid

Your brand voice, everywhere.

Clone one voice and use it across every market you serve.

brand-voice.wav0:06
Speaks 73 languages
How cloning works
Voices of Africa

1.1 billion speakers. One voice platform.

Our languages are spoken by more than a billion people. Turn on sound and each face greets you in turn.

Photos are illustrative. Every greeting is spoken by a STORM voice.

Solutions

Built for every industry that talks to its customers.

Keep scrolling to see each industry, and press play to hear it.

Banking and fintech

Talk to customers in the language they bank in.

  • Transaction alerts and balance updates read aloud
  • IVR and call-centre menus in local languages
  • Onboarding and KYC guidance
Bank alertsYoruba · John

“Ẹ kú àṣeyọrí! Ẹ ti gba twenty thousand naira sínú àkáǹtì yín. Owó tó kù nínú àkáǹtì yín báyìí jẹ́ one hundred and eighty-five thousand, five hundred naira.”

In English: Congratulations! You have received ₦20,000. Your balance is now ₦185,500.

Play
Languages

73 languages. 44 countries.

One integration covers languages from Acholi to Zulu, so a new market is a parameter, not a project.

languages to speak
73
TTS
languages to hear
5
ASR
languages to translate
5
MT
Countries
44
Speakers
1.1B+
Selected on the globe
YorùbáYoruba
Nigeria, BeninText to speech, voice cloning, speech to text and translation

Drag to turn the globe. Select a language to see what it supports.

All 73 languages

Native voices, speech to text and translationText to speech and voice cloning

For developers

From text to voice in a few lines.

Plain REST and JSON
No SDK to install and no model to host.
Examples for every endpoint
cURL, Python and Node, plus the whole guide as Markdown.
Safe key rotation
Old keys keep working through a grace window while you deploy.
Usage on record
Requests, characters, audio minutes and cost by service, model and language.
POST /v1/speech
import requests

r = requests.post(
    "https://api.plotweaver.app/v1/speech",
    headers={"X-API-Key": API_KEY},
    json={
        "text": "Your transfer of ₦25,000 was successful.",
        "language": "pcm",
        "voice_id": "efosa",
    },
)
open("alert.wav", "wb").write(r.content)

# Spoken as “twenty-five thousand naira”
Get started

From request to first audio. Four steps, nothing to host.

Every account is set up by our team, so there are no open sign-ups. Here is how it works.

  1. 01

    Request access

    Tell us what you are building and the languages you need.

    Request access
  2. 02

    We set up your account

    We switch on your services, and you can start a free trial of 100 minutes.

  3. 03

    Create an API key

    Sign in to the dashboard and create a key for each environment.

  4. 04

    Make your first call

    Send text to POST /v1/speech and get natural audio back.

Enterprise

Built for teams that answer to compliance.

Managed accounts

We create your team’s accounts and switch on only the services you’ve agreed. There are no open sign-ups.

Private by default

Cloned voice clips are stored privately and only served through short-lived signed links.

Usage you can audit

Every request is logged by service, model and language, with PDF usage reports for finance.

Keys you control

Create, expire, rotate and revoke API keys from the dashboard without touching production.

FAQ

Questions, answered.

Everything teams usually ask before their first call.

Ask us something else
What is the PlotWeaver Developer API?

One API for text to speech, voice cloning, speech to text and translation in African languages. You send text or audio and get speech, transcripts or translations back.

Who is it for?

Businesses that talk to customers in African languages: banks and fintechs, telecoms, media and advertising, health, education, logistics and the public sector, and the developers who build for them.

Which languages are supported?

Text to speech and voice cloning cover 73 languages on STORM v3. Yoruba, Hausa, Igbo, Nigerian Pidgin and Nigerian English also have hand-tested native voices on STORM v1, translation, and speech to text on STORM ASR 2.0, which also understands code-mixed speech such as Yoruba with English.

How do I get access?

Request access and tell us what you are building. We create your organisation’s account and switch on the services you need. There are no open sign-ups.

How does billing work?

Every account can start with 100 free minutes over 7 days. Then pay by card through Stripe: top up and pay as you go at $0.06 a minute, or choose Starter, Growth or Business for a monthly pool of minutes. Full details are on the pricing page.

Can we try it before we commit?

Yes. Every new account gets a one-time $2 credit to test the voices on your own content, in the Playground or with your API keys, before choosing a plan.

What happens to our cloned voices?

Reference clips are stored privately and only served through short-lived signed links. Cloned voices belong to your account and only work with your API keys.

Start building

Your product is one API call from every voice your market speaks.

Tell us what you’re building. We’ll set up your account and switch on the services your team needs.