VoicePay

Make Mobile Money Transactions with Your Voice.

Comment

Inspiration

In Uganda, mobile money is a lifeline for financial transactions, yet many users face barriers such as literacy challenges, visual impairments, and the complexity of USSD menus. Observing these struggles, we envisioned a solution that empowers every individual to access mobile money services effortlessly. VoicePay was born out of a desire to create a more inclusive financial system using cutting-edge voice recognition technology.

What it does

VoicePay enables users to perform mobile money transactions using simple voice commands in their preferred local languages.

  • Send Money: Transfer funds securely using just your voice.
  • Check Balance: Instantly know your account balance.
  • Accessibility: Designed for users with limited literacy or visual impairments.
    With VoicePay, financial transactions become faster, easier, and more accessible than ever.

How we built it

  1. Frontend:

    • Built with React Native for cross-platform compatibility (iOS and Android).
    • Integrated @react-native-community/voice for voice recognition.

  2. Backend:

    • Deployed on AWS Lambda for scalable, serverless computing.
    • Used AWS Transcribe to process voice inputs into text.
    • Implemented AWS Lex to detect intents like sending money or checking balance.
    • Provided voice feedback via AWS Polly.

  3. Security:

    • Integrated AWS Cognito for user authentication.
    • Used AWS Key Management Service (KMS) for encrypting sensitive data.

  4. Mock Database:

    • Simulated user accounts and balances using AWS DynamoDB.

Challenges we ran into

  • Multilingual Voice Recognition: Training models to accurately understand local languages like Luganda and Runyankore required fine-tuning and collecting representative datasets.
  • USSD Integration: Simulating USSD responses in real-time was complex, given the lack of direct APIs from telecom operators.
  • User Feedback: Balancing simplicity with functionality for users of diverse tech literacy levels was a continuous process.
  • Offline Support: Ensuring core functionality works without an internet connection required edge computing solutions.

Accomplishments that we're proud of

  • Successfully implemented a working prototype that can handle basic voice commands.
  • Integrated multiple AWS services seamlessly for voice recognition and processing.
  • Designed a clean and intuitive user interface that caters to users with various needs.
  • Created a scalable architecture ready for future expansions and integrations.

What we learned

  • The importance of accessibility in financial services and how technology can bridge gaps.
  • Deep insights into voice processing and natural language understanding, especially for underrepresented languages.
  • Practical experience with AWS services like Lambda, Lex, Polly, and Transcribe.
  • Overcoming technical and logistical challenges in building a voice-first application.

What's next for VoicePay

  1. Multilingual Support: Expand to more Ugandan languages like Swahili and Acholi.
  2. Direct API Integrations: Partner with telecom operators to streamline USSD interactions.
  3. Offline Mode: Enhance offline capabilities with AWS IoT Greengrass for processing on edge devices.
  4. Financial Literacy Programs: Embed tutorials to educate users on managing finances effectively.
  5. User Feedback Loop: Continuously improve language models and UI based on user input.

VoicePay aims to redefine how Ugandans interact with mobile money, making financial independence achievable for all.

Built With

Share this project:

Updates