Giving emotional identity to text.
Sono is a lightweight prototype designed to support people — especially neurodivergent individuals — who find it difficult to interpret emotional tone in spoken or written communication. When speech becomes text, emotional context is often lost. Sono explores how subtle visual cues can help restore that context, making communication clearer, gentler, and more accessible.
- Live transcription UI with emotionally expressive text
- Subtle visual cues (color, emphasis, optional emojis) to suggest emotional tone
- User-controlled toggles to adjust how much emotional information is shown
- Light and dark mode support
- Calm, minimal interface designed to reduce cognitive overload
Sono does not label, score, or diagnose emotions. Instead, it offers optional signals that can help users better interpret how something may be expressed.
- Frontend: React + Tailwind CSS
- Design focus: accessibility, emotional clarity, and user agency
- Backend: Placeholder for real-time speech-to-text and emotion inference
This repository currently focuses on the frontend experience and visual language. Backend integration is intentionally left open for future exploration.
Many neurodivergent people — including autistic individuals, people with ADHD, and others — experience difficulty interpreting tone, sarcasm, or emotional shifts in conversation. Traditional transcription tools remove even more context, making communication harder rather than easier.
Sono reframes transcription as a form of supportive translation, helping text retain emotional identity without forcing interpretation or judgment.
- Speech-to-speech exploration using emphasis and prosody models (placeholder)
- Emotionally expressive subtitles and overlays for calls, video, or presentations
- Expanded emotional nuance beyond basic categories
- User testing with neurodivergent communities to guide design decisions
npm install
npm run dev