Hey Gigi!

We understand that studying can be tough, especially when you have a short attention span or are struggling to focus. That’s why we built this app: learn like you’re "spilling the tea" with friends!

Comment

Inspiration

The inspiration for "Hey Gigi!" stems from the playful, "bestie" vibes of shows like Gossip Girl, aiming to create a friendly, approachable atmosphere, while incorporating a more studious and thoughtful atmosphere of shows like Gilmore Girls. We also realized how difficult it can be to stay engaged and study effectively, particularly for people with short attention spans or ADHD. Additionally, women and gender minorities in tech often face intimidation and insecurity when voicing their questions or concerns. Recognizing these challenges, we saw an opportunity to create a space where everyone can feel comfortable asking questions and engaging without fear of judgment. Thus, "Hey Gigi!" was born — the perfect solution to bridge these gaps.

What it does

Hey Gigi! is a voice-based AI chatbot that uses Google Speech-to-Text, Gemini AI, and Google Text-to-Speech APIs to provide seamless voice and text interactions. It can answer questions, explain complex concepts, or have casual conversations and/or calls with users, offering both text and speech responses. The application is designed to support users with ADHD or short attention spans, and it ensures an engaging experience with personalized interactions.

_"Hey Gigi!" is all about being your go-to AI bestie! I explain complex stuff in a way that's easy to understand, like we're just gossiping over coffee. I try to be supportive, funny, and, most importantly, relatable. You need a star explained? I got you. You need a wink? 😉 Consider it done. You need an ear to listen? I'm here, darling!

XOXO, Gigi_

How we built it

We integrated multiple cutting-edge APIs into our application:

  • Gemini AI: Handles the processing of the input and generates contextually relevant responses.
  • Google Speech-to-Text: Converts the user's speech into text, which is then sent to the AI model.
  • Google Text-to-Speech: Converts the AI's text responses back into voice, creating a seamless conversation.
  • React: Powers the front-end interface, ensuring a smooth user experience for chat and voice interactions.

Challenges we ran into

  • API Integration: Integrating different APIs (Speech-to-Text, Gemini AI, Text-to-Speech) proved to be challenging, especially when ensuring they worked seamlessly together in a React environment.
  • Handling Multiple Input Types: Combining voice, text, and file inputs while maintaining an intuitive user interface was difficult to implement smoothly.

Accomplishments that we're proud of

  • We successfully integrated voice and text interactions into a single interface.
  • The app can handle real-time speech input and give audio output, making it more accessible.
  • We ensured that the AI, "Gigi," can respond naturally to a variety of conversational cues and follow up on previous interactions.

What we learned

  • API Handling: Gaining deeper knowledge of integrating third-party APIs, including managing their responses and handling edge cases.
  • User Experience: We learned how crucial user experience is when designing for accessibility, especially when building apps for people with ADHD or those requiring voice-based inputs.
  • Collaboration: Working in a collaborative team environment allowed us to tackle problems from different perspectives and find creative solutions.

What's next for Hey Gigi!

  • Improved Speech Recognition: We're working on improving accuracy, especially in noisy environments.
  • Broader Accessibility Features: Adding more customization options to accommodate other disabilities and preferences, such as adjustable speech speed and tone.
  • Voice-Activated Commands: We want to integrate more features that can be activated through voice commands to further improve accessibility and convenience.
Share this project:

Updates