Inspiration
We noticed how tedious and repetitive form-filling can be, especially for people on the move or with accessibility needs. With rising interest in voice interfaces and conversational AI, we envisioned a tool that replaces manual form entry with seamless voice input boosting productivity
What it does
AudEntra is a voice-powered form-filling platform. Users can speak naturally, and our AI transcribes and maps their responses directly into digital forms. It supports smart suggestions, team collaboration, and voice intent recognition making data entry fast, hands-free, and intuitive.
How we built it
React + TypeScript for the frontend Node.js + Express for the backend Google Cloud Speech-to-Text API for transcription PostgreSQL + Supabase for data handling Vite for fast builds and deployment-ready bundling
Challenges we ran into
Some of the major hurdles we faced included handling inconsistent audio formats, ensuring high transcription accuracy across accents, and building a system that could map free-form voice responses to structured form fields. Dealing with large audio payloads and API limits added technical complexity, and integrating everything to offer smooth, real-time feedback took careful coordination between the frontend and backend. and also ensuring the ai conversationally worked as intended across all use cases
Accomplishments that we're proud of
We’re proud to have developed a fully working MVP that offers voice-based form filling along with team collaboration features. Our transcription pipeline not only works across various formats but also includes confidence scoring. The user interface prioritizes accessibility, and we’ve structured the app in a way that makes it easy to scale and deploy, which has been critical for testing and iteration.
What we learned
Throughout development, we gained deep insights into using speech APIs in production environments. We learned how to handle audio encoding and MIME types effectively, and explored the nuances of user experience design for voice-based interactions. It also taught us how to build an accessible system that balances automation with user control.
What's next for AudEntra
Looking forward, we plan to add multi-language support and speaker diarization for shared environments. We also aim to integrate with tools like Notion, Airtable, and Zapier to extend AudEntra’s utility. A desktop and mobile app version with offline transcription is on the roadmap, along with more advanced intent detection to automate even more complex form workflows.
Built With
- express.js
- google-cloud-speech
- node.js
- postgresql
- react
- supabase
- typescript
- vite
Log in or sign up for Devpost to join the conversation.