Vocal Well

Your Voice. Your Health. Your Power

Comment

VocalWell

Your Voice. Your Health. Your Power.

💡 Why It Matters

Over 500 million people suffer from voice disorders. Most don’t even know it — until it’s too late.

Teachers, singers, customer service reps, the elderly — and especially women in underserved regions — are at risk. VocalWell makes vocal care fast, private, and accessible for everyone.

🌟 What is VocalWell?

VocalWell is a simple, powerful web app that helps people detect voice disorders early — just by speaking into their phone.

No hospitals. No appointments. Just your voice, AI, and secure technology.

🩺 How It Works

  1. Upload or Record Your Voice (10 seconds)
  2. Our AI Model Analyzes It:
    • Uses BiLSTM + CNN
    • Extracts features like MFCCs, jitter, shimmer
    • Detects early signs of vocal issues (e.g., strain, nodules, tremors)
  3. Get Your Instant Health Report
    • Voice risk score
    • Pitch & waveform graphs
    • Detailed insights & next steps

🤖 Smart, Multilingual AI Helpline

  • Built using Langchain + open-source LLMs
  • Supports 10+ languages (English, Hindi, Punjabi, Tamil, Arabic, and more)
  • Offers:
    • Mental health support
    • Vocal therapy tips
    • Local specialist suggestions

🔐 Web3-Powered Privacy

  • Your voice report is encrypted and stored on IPFS
  • You get a private access link — only you control it
  • No centralized servers. No data leaks.

🛠️ Built With

  • AI/ML: TensorFlow, Keras, Librosa, NumPy, SciPy
  • Frontend: Next.js, Tailwind CSS, Framer Motion
  • Backend: Flask, PythonAnywhere
  • Storage: IPFS via NFT.storage
  • Agentic AI: Langchain + open-source LLMs

🏆 Why It’s Awesome

  • Full-stack MVP built in 48 hours
  • Built our own deep learning voice model
  • Inclusive: Works across languages and regions
  • Emotionally intelligent AI — not just a chatbot, it’s a carebot

📈 What’s Next

  • Support for 50+ global accents & dialects
  • WhatsApp/IVR version for rural access
  • Add detection for stress, respiratory issues, Parkinson’s
  • Collaborations with public health orgs

🌍 Impact

VocalWell gives people their voices back — literally and figuratively.

It fights:

  • Healthcare inaccessibility
  • Neglect of voice disorders
  • Mental health stigma
  • Language and data discrimination

👥 Team VocalWell

Built with love for voices that deserve to be heard.

🛠️ Built with:
css · flask · framer-motion · google-colab · html · ipfs · javascript · keras · langchain · librosa · next.js · nft.storage · numpy · open-source-llms · python · pythonanywhere · scipy · tailwind-css · tensorflow

Built With

Share this project:

Updates

posted an update —

VocalWell Just Got Smarter!

We’ve enhanced our voice health tech with AI-based anomaly detection to spot vocal irregularities before symptoms appear. A context-aware AI now interacts with users, answers health queries, and mimics real consultations.

We’ve also integrated Web3 + IPFS to ensure your medical reports are encrypted, decentralized, and immutable, giving you full control over your data. Reports are enriched with waveform graphs, MFCC plots, and deep diagnosis insights.

VocalWell is evolving into a complete vocal health ecosystem, now supporting regional languages and accents for a more inclusive experience. More updates are coming soon. This is just the beginning!

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