Video Analyser

Part of the analyser family.

A video analysis engine that extracts signals from presentations — speech transcription, delivery metrics, scene detection, and visual analysis — as structured JSON. (Rubric grading / assessment is intentionally not here: it's assessment-aware, so it lives in a lens above the analysers, not in this signal generator.)

Migrating from --grade (≤0.9.x)? Rubric-based assessment moved to assessment-lens (pip install assessment-lens), which maps signals from this analyser (and the rest of the family) to a rubric as observations, not grades — a human assigns every mark. video-analyser 0.10.0 is signal-only.

Status: Phase 1 MVP in development. Core infrastructure complete, video processing pipeline in progress.

Features

• Video Processing: Support for MP4, MOV, AVI, and WebM formats
• Speech Analysis: Automatic transcription with speaking rate and filler word detection
• Visual Analysis: Scene detection with frame captioning and quality assessment
• Structured Signals: Every metric emitted as JSON, ready for downstream assessment tools
• Professional Reports: Interactive HTML and structured JSON outputs

Installation

Prerequisites

• Python 3.11 or higher
• ffmpeg (for video processing)

Option 1: Install from PyPI (recommended for users)

pip install video-analyser

Option 2: Install from source (for development)

# Install uv (fast Python package manager)
curl -LsSf https://astral.sh/uv/install.sh | sh

# Clone the repository
git clone https://github.com/michael-borck/video-analyser.git
cd video-analyser

# Create virtual environment and install
uv venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
uv pip install -e ".[dev]"

Installing ffmpeg

macOS:

brew install ffmpeg

Ubuntu/Debian:

sudo apt update && sudo apt install ffmpeg

Windows: Download from https://ffmpeg.org/download.html

Quick Start

# Show available commands
video-analyser --help

# Check version
video-analyser version

# Launch web interface (coming soon)
video-analyser analyze

# Analyze a specific video (CLI mode - coming soon)
video-analyser analyze video.mp4 --output ./reports

Current Status: The CLI framework is complete. Video processing features are in active development.

Development

This project uses modern Python tooling and follows strict quality standards:

• uv for fast package management
• ruff for formatting and linting
• basedpyright for strict type checking
• pytest for testing with coverage
• pyproject.toml for all configuration (no setup.py)

Development Setup

# Clone and setup
git clone https://github.com/michael-borck/video-analyser.git
cd video-analyser
uv venv && source .venv/bin/activate
uv pip install -e ".[dev]"

# Verify setup
video-analyser --help
pytest -v

Code Quality Standards

# Format code
ruff format .

# Lint code  
ruff check .

# Type checking (strict mode)
basedpyright

# Run tests with coverage
pytest -v

# Run all quality checks
ruff format . && ruff check . && basedpyright && pytest -v

Project Structure

src/video_analyser/          # Main package
├── core/                # Video processing pipeline
├── analysis/            # Speech and visual analysis
├── reports/             # Report generation
├── interface/           # Gradio web interface
└── utils/               # Configuration and utilities

tests/                   # Test suite (mirrors src structure)
docs/                    # Documentation and specs
tasks/                   # Development task tracking
config/                  # Configuration files

Current Development Phase

• ✅ Phase 0: Project setup, packaging, PyPI publication
• 🚧 Phase 1: Core video processing pipeline (in progress)
• 📋 Phase 2: Enhanced analysis features
• 📋 Phase 3: Advanced AI features

See tasks/tasks-prd-phase1-mvp.md for detailed task tracking.

Links

• PyPI: https://pypi.org/project/video-analyser/
• GitHub: https://github.com/michael-borck/video-analyser
• Documentation: Coming soon

License

MIT License - see LICENSE file for details.

Contributing

Contributions are welcome! Please read the development guidelines in CLAUDE.md for our coding standards and toolchain requirements.