EmoRead: Using AI, Unlike AI

🧠 Inspiration

We wanted to address the declining interest in reading among students.

In today’s world, many students use AI tools to summarize texts instead of engaging deeply with them. Our goal was to bring back emotion, reflection, and connection into reading.

This inspired EmoRead — an AI-powered platform that transforms a reader’s emotions into dynamic color visuals and emotion graphs, making reading immersive and personal.


🎭 What EmoRead Does

EmoRead listens as the user reads aloud and detects emotions based on vocal features such as tone, pitch, and rhythm.

Emotion → Color Mapping

Emotion Color Code Color
Happy #22C55E Green
Sad #3B82F6 Blue
Angry #EF4444 Red
Neutral #6B7280 Gray
Fear #8B5CF6 Purple
Disgust #F97316 Orange
Confusion #14B8A6 Teal

As the user continues reading, the screen smoothly transitions between these colors to represent emotional flow.

End of Reading Session

  • A unique Emotion Graph visualizes the entire emotional journey.
  • A summary report highlights dominant emotions and their intensities.

Community Discussion

Readers can:

  • Share their emotion graphs
  • Compare interpretations
  • Reflect on emotional responses
  • Understand how peers felt about the same story

This makes reading interactive, expressive, and meaningful.


🛠️ How We Built It

Tech Stack

  • Frontend: React, TypeScript
  • Styling: Tailwind CSS
  • Core AI: Google Gemini API, SpeechBrain (SER Model)
  • Visualization: Recharts

We combined Speech Emotion Recognition with modern visual design to create a seamless reading–emotion experience.


⚠️ Challenges

  • Ensuring emotional accuracy across accents and voice patterns
  • Creating smooth, meaningful color transitions
  • Keeping latency low between emotion detection and UI updates
  • Designing a student-friendly, discussion-based interface

🏆 Accomplishments

  • Real-time speech-to-emotion color mapping
  • Visualizing emotional flow through dynamic transitions
  • Introducing emotional intelligence into reading — a rarely explored area
  • Making reading more engaging and reflective

📚 What We Learned

  • Integrating AI emotion models with real-time UI feedback
  • The huge role of user engagement in ed-tech
  • Effective teamwork between API, frontend, and data layers
  • Understanding that reading is emotional, not just cognitive

🚀 What’s Next

  • Emotion-based book sorting (e.g., calming, thrilling, relaxing reads)
  • Teacher analytics dashboard for emotional engagement tracking
  • Multimodal emotion detection combining voice + face + text

💡 Final Thought

EmoRead turns reading from a passive task into an interactive emotional journey.

By blending emotion, color, and storytelling, we aim to make reading joyful, reflective, and deeply human again.

Built With

Share this project:

Updates

posted an update

Project Update: EmoRead Moves Into Research Phase

We're excited to share that EmoRead is now progressing into a research and further development project!

Our next steps include improving speech emotion accuracy, refining emotion–color visualizations, and exploring real-world educational and reader-focused applications.

More updates soon as we continue building EmoRead beyond the hackathon version!

Log in or sign up for Devpost to join the conversation.