A Python-based voice analysis tool that evaluates speech quality for interview readiness using audio signal processing.
This project records or uploads speech audio and analyzes key vocal traits such as confidence, calmness, clarity, and pacing, then provides a readiness score with visual feedback.
- 🎙️ Record voice directly in Google Colab
- 📂 Upload pre-recorded audio files
- 📊 Analyze interview-relevant vocal traits:
- Confidence (pitch stability & energy)
- Calmness (pauses & jitter)
- Clarity (spectral centroid)
- Pacing (estimated words per minute)
- 📈 Visualizations:
- Energy over time
- Clarity indicator
- Pitch distribution
- 🔊 Audio playback after recording
The system uses audio signal processing techniques via Librosa:
- Pitch tracking (
piptrack) - RMS energy analysis
- Spectral centroid for clarity
- Silence detection for pause analysis
- Onset detection for speech rate estimation
Each metric is normalized and scored on a scale of 0–10, then combined into an overall interview readiness percentage.
- Python
- Librosa
- NumPy
- SciPy
- Matplotlib
- IPython
- Google Colab
- JavaScript (for in-browser recording)
- Upload the notebook to Colab
- Run all cells
- Use the buttons to record or upload audio
- Click Analyze Recording
pip install -r requirements.txt