SlideStream 2.0 User Guide

Comprehensive guide for creating professional video presentations with SlideStream's AI-powered tools.

Quick Start
Installation
Configuration System
Creating Your First Video
Working with Providers
Advanced Workflows
Troubleshooting
Best Practices

Quick Start

1. Install SlideStream

# Install with all AI providers
pip install slide-stream[all-ai]

2. Create Configuration

slide-stream init

This creates a slidestream.yaml file in your current directory with example configuration.

3. Set Up API Keys

Edit your configuration file or set environment variables:

export OPENAI_API_KEY="your-openai-key"
export ELEVENLABS_API_KEY="your-elevenlabs-key"

4. Create Your First Video

slide-stream create presentation.md output.mp4

Installation

Core Installation

pip install slide-stream

With Specific Providers

# For OpenAI (DALL-E, GPT, TTS)
pip install slide-stream[openai]

# For ElevenLabs premium voices
pip install slide-stream[elevenlabs]

# For Google Gemini
pip install slide-stream[gemini]

# For Anthropic Claude
pip install slide-stream[claude]

# For Groq (fast inference)
pip install slide-stream[groq]

# All AI providers
pip install slide-stream[all-ai]

System Requirements

Python 3.10+
FFmpeg (for video processing)

Installing FFmpeg

macOS:

brew install ffmpeg

Ubuntu/Debian:

sudo apt update && sudo apt install ffmpeg

Windows: Download from FFmpeg website

Configuration System

SlideStream 2.0 uses YAML configuration files for maximum flexibility and maintainability.

Configuration Discovery

SlideStream searches for configuration in this order:

./slidestream.yaml (current directory)
~/.slidestream.yaml (home directory)
Built-in defaults

Basic Configuration

# slidestream.yaml
providers:
  llm:
    provider: openai
    model: gpt-4o-mini
    
  images:
    provider: dalle3
    fallback: text
    
  tts:
    provider: elevenlabs
    voice: rachel

api_keys:
  openai: "${OPENAI_API_KEY}"
  elevenlabs: "${ELEVENLABS_API_KEY}"

settings:
  video:
    resolution: [1920, 1080]
    fps: 24
  cleanup: true

Environment Variables

Use environment variables for secure API key management:

# OpenAI (for DALL-E 3, GPT, and OpenAI TTS)
export OPENAI_API_KEY="sk-..."

# ElevenLabs (for premium TTS)
export ELEVENLABS_API_KEY="..."

# Stock photo providers (optional)
export PEXELS_API_KEY="..."
export UNSPLASH_ACCESS_KEY="..."

# Other LLM providers
export GEMINI_API_KEY="..."
export ANTHROPIC_API_KEY="..."
export GROQ_API_KEY="..."

Configuration Profiles

Create different configurations for different use cases:

Basic Profile (basic.yaml):

providers:
  llm:
    provider: none
  images:
    provider: text
  tts:
    provider: gtts

Professional Profile (pro.yaml):

providers:
  llm:
    provider: openai
    model: gpt-4o
  images:
    provider: dalle3
    fallback: pexels
  tts:
    provider: elevenlabs
    voice: rachel

api_keys:
  openai: "${OPENAI_API_KEY}"
  elevenlabs: "${ELEVENLABS_API_KEY}"
  pexels: "${PEXELS_API_KEY}"

Use with:

slide-stream create --config pro.yaml presentation.md video.mp4

Creating Your First Video

From Markdown

Create a simple Markdown file:

# Welcome to SlideStream

- Create professional video presentations
- Use AI to enhance your content
- Generate videos automatically

# Key Features

- AI-powered image generation
- Premium text-to-speech voices
- Smart content enhancement
- Professional video output

# Getting Started

- Install SlideStream
- Configure your providers
- Create your first video

Generate the video:

slide-stream create presentation.md my-video.mp4

From PowerPoint

SlideStream 2.0 supports PowerPoint files with enhanced features:

slide-stream create slides.pptx presentation.mp4

PowerPoint Features:

Extracts slide titles and content
Uses speaker notes for enhanced narration
Preserves slide structure
Supports complex layouts

Example Speaker Notes

When using PowerPoint files, add speaker notes for better AI narration:

Slide Content: Key Benefits
• Faster development
• Better user experience
• Competitive advantage

Speaker Notes: In this slide, we'll explore the three main benefits of adopting our solution. First, you'll see dramatically faster development cycles, allowing your team to ship features in weeks rather than months. Second, your users will experience a more intuitive and responsive interface. Finally, these improvements will give you a significant competitive advantage in your market.

The AI will use these notes to create natural, flowing narration.

Working with Providers

Image Providers

DALL-E 3 (AI Generation)

providers:
  images:
    provider: dalle3
    fallback: text

api_keys:
  openai: "${OPENAI_API_KEY}"

Benefits:

Custom images for each slide
Relevant to your content
Professional quality
No licensing concerns

Requirements:

OpenAI API key
Pay-per-image pricing

Stock Photo Providers

Pexels:

providers:
  images:
    provider: pexels
    fallback: text

api_keys:
  pexels: "${PEXELS_API_KEY}"

Unsplash:

providers:
  images:
    provider: unsplash
    fallback: text

api_keys:
  unsplash: "${UNSPLASH_ACCESS_KEY}"

Text-Based Images

providers:
  images:
    provider: text

Always available as a fallback. Creates clean, professional text-based slides.

Text-to-Speech Providers

ElevenLabs (Premium)

providers:
  tts:
    provider: elevenlabs
    voice: rachel  # or adam, aria, etc.

api_keys:
  elevenlabs: "${ELEVENLABS_API_KEY}"

Available Voices:

rachel: Professional female voice
adam: Clear male voice
aria: Expressive female voice
josh: Warm male voice
And 900+ more voices

OpenAI TTS

providers:
  tts:
    provider: openai
    voice: nova  # alloy, echo, fable, nova, onyx, shimmer

api_keys:
  openai: "${OPENAI_API_KEY}"

Google TTS (Free)

providers:
  tts:
    provider: gtts

Always available, no API key required.

LLM Providers

Content Enhancement

LLMs improve your slide content by:

Making bullet points flow naturally
Creating engaging narratives
Improving clarity and structure
Generating better image search queries

OpenAI GPT:

providers:
  llm:
    provider: openai
    model: gpt-4o-mini  # or gpt-4o for higher quality

Google Gemini:

providers:
  llm:
    provider: gemini
    model: gemini-1.5-flash

Anthropic Claude:

providers:
  llm:
    provider: claude
    model: claude-3-5-sonnet-20241022

Advanced Workflows

Multi-Configuration Workflow

Create different configurations for different scenarios:

# Quick prototype with free services
slide-stream create --config basic.yaml draft.md prototype.mp4

# High-quality final version
slide-stream create --config premium.yaml final.md presentation.mp4

# Client-specific branding
slide-stream create --config client-brand.yaml proposal.md client-video.mp4

Batch Processing

Process multiple presentations:

# Create multiple videos
for file in *.md; do
    output="${file%.md}.mp4"
    slide-stream create "$file" "$output"
done

Custom Video Settings

Fine-tune video output:

settings:
  video:
    resolution: [1920, 1080]  # 4K: [3840, 2160]
    fps: 30                   # Smooth playback
    codec: libx264            # Compatibility
    audio_codec: aac
    slide_duration_padding: 2.0  # More time per slide
    default_slide_duration: 8.0

  image:
    bg_color: "#1a1a1a"      # Dark theme
    font_color: "#ffffff"
    title_font_size: 120
    content_font_size: 80

Integration with CI/CD

Automate presentation generation:

# .github/workflows/presentations.yml
name: Generate Presentations
on:
  push:
    paths: ['presentations/*.md']

jobs:
  generate:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-python@v4
        with:
          python-version: '3.11'
      
      - name: Install SlideStream
        run: pip install slide-stream[all-ai]
      
      - name: Generate Videos
        env:
          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
          ELEVENLABS_API_KEY: ${{ secrets.ELEVENLABS_API_KEY }}
        run: |
          slide-stream create presentations/quarterly-review.md output/q4-review.mp4
          slide-stream create presentations/product-launch.md output/launch.mp4

Troubleshooting

Common Issues

"Provider not available"

slide-stream providers

Check which providers are available and their status.

"API key not found"

Verify environment variables: echo $OPENAI_API_KEY
Check configuration file syntax
Ensure API keys are valid and have sufficient credits

"FFmpeg not found"

Install FFmpeg:

# macOS
brew install ffmpeg

# Ubuntu/Debian
sudo apt install ffmpeg

Video generation fails

Check disk space in temp directory
Verify input file format (.md or .pptx)
Try with --config to use specific configuration

Audio/video sync issues

Adjust timing settings:

settings:
  video:
    slide_duration_padding: 2.0  # More padding
    default_slide_duration: 6.0  # Longer default

Debug Mode

For detailed error information:

PYTHONPATH=. python -m slide_stream.cli create --help

Provider Status

Check what's working:

slide-stream providers

Output shows each provider's availability and requirements.

Best Practices

Content Creation

Markdown Structure:

# Clear, Descriptive Titles

- Use bullet points for key ideas
- Keep points concise and focused
- Aim for 3-5 points per slide

# Logical Flow

- Structure your presentation logically
- Use consistent formatting
- Include call-to-action slides

PowerPoint Tips:

Use speaker notes for detailed explanations
Keep slide content brief
Use consistent layouts
Include relevant images in slides

Configuration Management

Environment Variables:

# .env file (don't commit to git)
OPENAI_API_KEY=sk-...
ELEVENLABS_API_KEY=...
PEXELS_API_KEY=...

Version Control:

Commit configuration files
Use environment variables for secrets
Create different configs for different environments

Performance Optimization

Fast Generation:

providers:
  llm:
    provider: groq  # Fastest inference
    model: llama-3.1-8b-instant
  images:
    provider: text  # No API calls
  tts:
    provider: gtts  # Free and fast

High Quality:

providers:
  llm:
    provider: openai
    model: gpt-4o  # Best quality
  images:
    provider: dalle3
  tts:
    provider: elevenlabs
    voice: rachel

Cost Management

Monitor Usage:

OpenAI: Check usage dashboard
ElevenLabs: Monitor character usage
Set usage alerts

Optimize Costs:

Use text images for drafts
Switch to premium providers for final versions
Batch process multiple presentations

Quality Guidelines

For Professional Presentations:

Use DALL-E 3 or stock photos for images
Use ElevenLabs or OpenAI TTS for voices
Enable LLM content enhancement
Use higher resolution (1920x1080 or 4K)

For Internal/Draft Use:

Text-based images are sufficient
gTTS provides adequate quality
Disable LLM enhancement for speed

Security Best Practices

Never commit API keys to version control
Use environment variables or secure vaults
Rotate API keys regularly
Monitor API usage for unusual activity
Use least-privilege API permissions

This guide covers the essential workflows for SlideStream 2.0. For development setup and contributing, see DEVELOPMENT_WORKFLOW.md.

FilesExpand file tree

USER_GUIDE.md

Latest commit

History