AI Integration
Future-ready for voice recognition and automated scoring
Our platform is architected to incorporate AI-based speech recognition for automatically scoring learners in local languages and dialects. While full AI integration is planned for Year 2, the system foundation is ready from day one.
Year 2 AI Roadmap
In partnership with specialist voice model developers, we will deliver:
Current AI-Ready Capabilities
Audio Capture
High-quality voice recording with noise suppression already built into assessment flow.
Multi-Language Architecture
System designed for any language without translation dependencies.
Extensible Design
API-ready architecture for plugging in external AI services.
Scoring Pipeline
Existing scoring logic can be augmented with AI-generated scores.
Partnership Approach
We will partner with an organisation selected in conjunction with the Binding Constraints Lab that is currently building high-functioning voice models for education applications.
Voice Data Collection
Support collection of voice samples in African Home Languages through fieldwork.
Permission Navigation
Manage consent and ethical requirements for voice data from children.
Data Cataloguing
Store and organize voice data appropriately for model training.
Future AI Capabilities
Beyond voice recognition, the platform architecture supports: