Inspiration

The inspiration for InstaNote came from the need to quickly capture fleeting thoughts and ideas without the hassle of typing or writing them down. We wanted to create an app that makes it effortless to convert spoken words into organized notes, leveraging AI to enhance accuracy and readability.

What it does

InstaNote allows users to record audio notes, transcribe them using AI, and store both the recordings and transcriptions securely in the cloud. Users can edit, refine, and share their notes, with premium features available for subscribed users.

How we built it

I built InstaNote using SwiftUI with MVVM architecture, integrated with Firebase for authentication and Firestore for the database. Speech-to-text and text enhancement features are powered by Google’s Gemini AI integrated in Firebase Vertex AI and OpenAi's GPT-4o mini, providing enhanced transcription services.

Challenges we ran into

  • Since Vertex AI is in preview state, finding correct documentation and handling errors was challenging. I had to explore Google Cloud and run features on the web to get the parameters I needed.
  • Security was also a key concern. InstaNote sends data directly using Firebase SDK without relying on a custom backend with API tokens, requiring careful implementation of Firebase AppCheck, Apple AppAttest, and DeviceCheck.

Accomplishments that we're proud of

I’m proud of creating a seamless user experience that combines advanced AI technology with a simple, intuitive interface. Successfully integrating transcription and enhancement using generative AI were key milestones for me.

What we learned

I learned how to integrate AI for speech-to-text processing and the importance of security with AppAttest and DeviceCheck.

What's next for InstaNote

Next, I plan to expand InstaNote by adding features like conversational AI, which will allow users to enhance their thoughts and notes further with the help of AI.

Built With

Share this project:

Updates