voiceOGraphy

voiceOGraphy is an AI-powered voice analysis web application that helps users understand and improve their speech delivery by converting raw audio into actionable feedback. Built in just 10 hours using the Trea AI IDE during the Code Craft AI x Dev Hackathon, voiceOGraphy provides real-time analysis of critical vocal parameters such as pace, pitch, volume, and clarity. Using audio signal processing techniques like Root Mean Square (RMS) for volume detection, Zero Crossing Rate (ZCR) for pitch estimation, Voice Activity Detection (VAD) for noise filtering, spectral centroid for tone brightness, and autocorrelation for pitch accuracy, the app translates low-level acoustic data into meaningful insights. For instance, it can label a speaker’s delivery as “fast pace,” “low pitch,” or “clear volume,” helping users—especially public speakers, educators, and content creators—understand their vocal strengths and areas for improvement. Our tech stack includes React.js, Vite, Tailwind CSS for the frontend, and Node.js, Express.js, Multer, and Axios on the backend. Future upgrades will integrate AssemblyAI for transcription and advanced voice intelligence. The entire system was developed collaboratively under time pressure, showcasing how fast innovation can happen with the right tools and teamwork.