VoiceU

Inspiration

30,000+ Americans lose their voice each year to ALS, cancer, or stroke. Professional voice banking costs $3,000+ and takes months, most people miss the window. We wanted to make voice preservation free, fast, and accessible to everyone, while also enabling creative use cases like content creation and family memories.

What it does

VoiceU clones your voice in 15 minutes: Record 8 guided prompts (childhood memory, life advice, common phrases) AI generates your voice model in 5-10 minutes Type anything and hear it in your voice Save phrases, download audio, share with family

How we built it

Frontend: Next.js 15 with TypeScript and Tailwind CSS Voice Recording: Web Audio API with real-time waveform visualization Voice Cloning: ElevenLabs API for AI voice synthesis Backend: Supabase (PostgreSQL + Storage) for user data and audio files Deployment: Vercel

Challenges we ran into

Audio Quality: Ensuring clean recordings across different microphones and environments Voice Accuracy: Optimizing prompts to capture emotional range and natural speech patterns Processing Time: Managing 5-10 minute voice generation with clear user feedback

Accomplishments that we're proud of

We built a fully functional voice cloning app that reduces a $3,000, months-long process into a free 15-minute experience.

What we learned

Voice cloning requires carefully designed prompts to capture emotional range and natural speech patterns, not just generic recordings.

What's next for VoiceU

We plan to add family sharing with QR codes, and integration with AAC devices for people who need it most. Long-term, we're exploring partnerships with ALS associations and cancer centers to make voice preservation a standard part of medical care.