examples

Example Outputs

This directory contains sample outputs from the audio extraction and analysis pipeline to help you understand what to expect from the tool.

Sample Output Files

When you run the tool, it generates outputs similar to these examples. Use --analysis-style concise (default) for a single markdown analysis, or --analysis-style full for the 5-file breakdown. Add --export-markdown on transcribe/process to also emit a formatted transcript.

📝 Text Output (--format text)

Simple, clean transcript without additional formatting or analysis.

📊 JSON Output (--format json)

Structured data including:

• Complete transcript
• Timestamps for each segment
• Speaker identification (when available)
• Metadata (duration, word count, etc.)

📄 Markdown Output (--format markdown)

Comprehensive analysis document containing:

1. Executive Summary - High-level overview of the content
2. Chapter Overview - Content broken into logical sections
3. Key Topics & Intents - Main themes and discussion points
4. Full Transcript - Complete text with timestamps
5. Key Insights & Takeaways - Actionable insights from the content

How to Generate Your Own Examples

Basic Example

audio-extraction-analysis process your-video.mp4 --format markdown

Complete Pipeline Example

audio-extraction-analysis process podcast.mp4 \
    --output-dir examples/my-podcast/ \
    --quality speech \
    --provider deepgram \
    --format markdown \
    --verbose

File Naming Convention

Generated files follow this pattern:

• {filename}_transcript.txt - Plain text transcript
• {filename}_transcript.json - JSON formatted data
• {filename}_analysis.md - Full markdown analysis
• {filename}_audio.mp3 - Extracted audio file

Privacy Note

The examples in this directory are for demonstration purposes. When processing your own content, outputs will be saved to data/output/ by default, which is git-ignored to protect your privacy.

Sample Content Types

This tool works well with:

• 🎙️ Podcasts - Interview style, multiple speakers
• 📺 YouTube Videos - Educational content, tutorials
• 🎬 Webinars - Presentations, Q&A sessions
• 🗣️ Speeches - Keynotes, lectures
• 📹 Meetings - Recorded video calls, conferences

Each content type may benefit from different quality settings:

• Use --quality speech for spoken content
• Use --quality music for content with background music
• Use --quality high for maximum fidelity