Record. Analyze. Improve. — Master your interview delivery with objective, data-driven feedback.
Technion CS 236272 · Project in Android Development · 2026
In today's competitive job market, soft skills are as crucial as technical knowledge. Many candidates fail interviews due to poor delivery — lack of eye contact, a nervous pace, or excessive filler words like "umm" and "like."
InterviewPro is a "Smart AI Mirror": you record a practice interview, and the app gives you objective, quantitative feedback on exactly how you came across — so you can improve with numbers, not guesses.
| 🎯 Feature | What it does | Status |
|---|---|---|
| 🔐 Private Accounts | Secure email/password login — sessions stay private and synced | ✅ |
| 🎥 Record & Replay | Capture practice interviews and replay them to hear your tone | ✅ |
| 📝 Full Transcript | AI speech-to-text (OpenAI Whisper) of everything you said | ✅ |
| 🗣️ Filler Word Detection | Counts every "um," "like," "you know" with a per-minute rate | ✅ |
| 📚 Vocabulary Analysis | Flags overused words and measures vocabulary diversity | ✅ |
| ⏸️ Silence Detection | Spots long, awkward pauses that break your flow | ✅ |
| 📊 Progress Comparison | Compares any two sessions side-by-side to track improvement | ✅ |
| 📈 Performance Report | A single score plus skill breakdown, stats, and a pace chart | ✅ |
Performance Report — overall score, skill breakdown, key metrics, and speaking-pace trend, all in one glance.
| # | User Story | Status |
|---|---|---|
| #6 | Private account for safe, synced practice data | ✅ |
| #9 | Save and replay recordings | ✅ |
| #10 | Read a text transcript of everything I said | ✅ |
| #3 | Detect and count filler words ("umm," "like") | ✅ |
| #4 | Flag over-repeated professional words | ✅ |
| #5 | Flag long or awkward silences | ✅ |
| #8 | Compare latest practice with a previous one | ✅ |
| # | User Story |
|---|---|
| #11 | AI coaching tips at session end, with rephrasing suggestions |
| #12 | Detect if I sound confident, nervous, or bored |
| #7 | Visual graphs of improvement over weeks |
| #14 | Feedback on hand gestures & body language |
| #15 | Visual map of where I was looking on screen |
| # | User Story |
|---|---|
| #16 | Emotional tone detection (nervous / happy / bored) |
| #17 | Detect if I scan the whole "audience" vs. one fixed point |
| #18 | Hand-gesture analysis for more engaging body language |
Flutter 3.38.3 · Dart · Firebase (Auth + Firestore) · OpenAI Whisper · fl_chart · Provider
git clone https://github.com/Technion236272/2026b-InterviewPro.git
cd 2026b-InterviewPro
fvm use 3.38.3
fvm flutter pub get
flutterfire configure # connect your own Firebase
echo "OPENAI_API_KEY=sk-your-key" > .env # add your Whisper key
fvm flutter runBuilt with Flutter 💙 · Powered by OpenAI Whisper