ASR
Automatic Speech Recognition
Transcribe Akan Twi with English code-switching, Twi-only speech, Ghanaian Pidgin, and English from a microphone or audio file.
Open ASR Playground
Speech APIs & on-device SDKs
Abena AI
African-language AI, from voice to video.
Build with automatic speech recognition (ASR) and text-to-speech (TTS) for African languages and regional accents. Integrate hosted APIs or deploy on-device SDKs.
For products, teams, and creators
Speech infrastructure. Creative possibilities.
Add voice to customer service, education, and business workflows. Build with our APIs and SDKs, or explore voice and video tools in the app.
TTS
Text to Speech
Turn text into natural speech with supported African-language, regional English, and Pidgin voices through the web or API.
Open TTS PlaygroundSDKs
On-Device Speech
Run supported speech models on your own devices. Enable offline experiences, keep speech data local, and reduce cloud dependency.
Explore On-Device SDKsCreative AI
Multilingual Video Creation
Create music videos, studio sessions, video podcasts, drama, and dubbed content using African and global voices in the mobile app.
Create on AndroidHosted speech APIs
From evaluation to integration
Evaluate models in the browser, then use the documented TTS and ASR endpoints from Python, JavaScript, mobile apps, or your existing backend. Add transcription, spoken responses, and accessible content without operating your own speech servers.
POST
/playground/api/v1/tts/synthesize/
POST
/playground/api/v1/asr/transcribe/
On-device SDKs for enterprise
Your product. Your device. Your control.
Deploy supported ASR and TTS models on your hardware. After setup, speech processing runs locally, without an internet connection.
Model availability and hardware requirements vary by deployment. Our hosted APIs and browser playgrounds process requests online.
Discuss SDK Integration
Keep speech data local
Process audio and text on the device without sending them to a speech server.
Work beyond connectivity
Keep voice workflows available offline and remove the network round trip from speech processing.
Reduce cloud costs
Use device compute to reduce server-side inference and audio-transfer costs. Contact us for SDK licensing.
Abena AI mobile app
One app for voice, smart home, and local-language creation
Abena is designed for daily use, not only demonstrations. Speak to your phone, control compatible smart-home devices, complete common tasks, and create media without leaving the app.
- Hands-free phone and app control
- Smart-home voice commands
- Mobile money, airtime, and bill actions
- Calls, messages, news, weather, and voice notes
- Music videos, podcasts, studio sessions, and drama
Proprietary data
Built with native speakers and real recording workflows
Our team works with voice artists, language reviewers, and audio engineers to create and evaluate speech data for languages often underserved by mainstream technology.
- Native-speaker studio recording
- Language review and quality control
- Training and evaluation across real use cases
Inside our speech-data recording process
Recognition
Recognized for African-language innovation
Explore independent coverage and international recognition of Abena AI.
Where Abena fits
Language technology for real products and services
Customer experience
Local-language support, voice agents, and call transcription.
Education and accessibility
Spoken learning content, literacy tools, and accessible interfaces.
Finance and field work
Hands-free workflows, voice-enabled services, and local-language guidance.
Media and entertainment
Natural voices, podcasts, music videos, drama, and localized content.
Work with Abena AI
Build the next African-language experience
Start in the playground or contact us about APIs, on-device deployment, data, and partnerships.