Cree Learner

Inspiration

Indigenous languages like Cree are disappearing at an alarming rate, and learning materials are often limited or inaccessible. We realized that there are many streets in Vancouver which have indigenous names that we couldn't read or pronounce. We developed a natural curiosity for the various languages and cultures of the indigenous peoples living with us in Vancouver today. We wanted to create a simple, mobile-friendly app that helps people learn and practice Cree pronunciation using AI feedback. Inspired by the idea of preserving Indigenous culture through technology, we decided to build a tool that can evaluate pronunciation and encourage learners to improve.

What it does

Our app allows users to:

Learn through interactive letter and sound cards, starting with Cree vowels and consonants.
Listen to correct Cree pronunciations.
Record their own voice and receive AI-based pronunciation feedback using Azure Speech Service.
View accuracy scores.
Support the indigenous communities. The app aims to make Indigenous language learning accessible, interactive, and culturally respectful.

How we built it

Frontend: React Native for iOS/Android app development.
Backend: Flask server running on Python, handling API requests between the app and Azure Speech Service, our Azure Database.
AI Integration: Azure Pronunciation Assessment API for real-time feedback.
Testing: iPhone device testing via Xcode, connected to the same Wi-Fi network for local Flask API calls.
Collaboration: GitHub, Notion and Figma for collaboration and project management.

Challenges we ran into

Device-Server Connectivity: Making the Flask backend accessible from iPhone devices on the same Wi-Fi network was tricky due to IP settings.
Xcode Build Errors: Debugging iOS build errors like exit code 65 and developer profile trust issues.
Azure Integration: Properly formatting audio data and managing API responses from Azure Speech.
Android Deployment: Troubling deployment using Android emulator.

Accomplishments that we're proud of

Created a meaningful tool that supports Indigenous language revitalization efforts.
Established an online database to store words and audio files and implemented API requests so that it could access this data in real-time
Successfully connected the React Native frontend to the Flask backend and received real-time pronunciation feedback.
Built a working prototype that records, evaluates, and displays AI feedback.
Learned to integrate Azure Speech with mobile applications.

What we learned

How to integrate AI speech services into mobile apps.
The importance of cross-device network setup when testing locally hosted servers.
How to debug iOS builds and manage developer certificates.
That combining technology and cultural preservation can create powerful and inclusive educational tools.