AI interview coach

session report
home page
live session

Inspiration

Public speaking is one of the most important skills for students, yet most people don’t get real-time feedback while practicing. We wanted to build an AI coach that listens, analyzes, and helps users improve instantly.

What it does

This project is an AI-powered speaking coach that:

Captures speech using browser microphone
Converts speech to text using Web Speech API
Analyzes speaking performance using Gemini AI
Generates feedback on clarity, confidence, filler words, and speed
Creates a post-session performance report

How we built it

Frontend built using HTML, CSS, JavaScript
AI feedback powered by Gemini API
Speech recognition using Browser Web Speech API
Data handling and session flow designed for real-time interaction
Deployment via Lovable + GitHub ## Challenges we ran into
Real-time transcription stability
Speech recognition accuracy
AI feedback pipeline timing
UI syncing with live metrics
Managing API usage and deployment limits during the hackathon ## Accomplishments that we're proud of
Built a real-time AI speaking assistant from scratch during the hackathon
Successfully integrated speech-to-text with live session tracking
Implemented AI-based feedback using Gemini
Created a performance report system for post-session analysis
Designed a clean and interactive UI for better user experience
Managed deployment and GitHub integration under time constraints
Solved real-world problem: helping students practice speaking confidently with instant feedback
## What we learned
Building real-time AI apps
Speech-to-text pipelines
Prompt engineering for feedback systems
Handling live session metrics
Deploying and presenting production-ready prototypes under time pressure ## What's next for AI interview coach
Emotion detection from voice
Interview practice mode
Presentation rehearsal scoring
Personalized improvement tracking
Mobile support