Skip to main content

​
Overview

The AI Video Presentation Generator requires proper configuration of environment variables and directory structure to function correctly. This guide covers all necessary setup steps for both backend and frontend.

​
Environment Variables

All environment variables should be configured in a .env file located in the backend/ directory.

​
Required API Keys

​
GEMINI_API_KEY
string
required
Google Gemini API key for AI-powered content generation. Used by the ContentGenerator and ScriptGenerator to create presentation content and narration scripts.See API Keys for setup instructions.
​
SARVAM_API_KEY
string
required
Sarvam AI API key for multi-language text-to-speech voice generation. Required for generating audio narration in supported languages.See API Keys for setup instructions.
​
UNSPLASH_ACCESS_KEY
string
required
Unsplash API access key for fetching high-quality images for presentation slides. Used when slides require visual content.See API Keys for setup instructions.

​
TTS Configuration

​
SARVAM_TTS_URL
string
default:"https://api.sarvam.ai/text-to-speech"
Sarvam AI text-to-speech API endpoint URL. Typically does not need to be changed.
​
SARVAM_MODEL
string
default:"bulbul:v2"
Sarvam AI TTS model version. Valid options:
  • bulbul:v2 (recommended, stable)
  • bulbul:v3-beta (latest features, may be unstable)

​
Server Configuration

​
HOST
string
default:"0.0.0.0"
Backend server host address. Use 0.0.0.0 to allow connections from any network interface.
​
PORT
number
default:"8000"
Backend server port number. Default is 8000 for FastAPI.

​
Directory Structure

The backend automatically creates the following directory structure in backend/outputs/:

​
Output Directories

backend/outputs/
├── scripts/        # Generated narration scripts (JSON)
├── slides/         # Slide content data (JSON)
├── manim_code/     # Generated Manim animation code (Python)
├── videos/         # Rendered Manim animations (MP4)
├── audio/          # Generated voice narration (WAV)
├── final/          # Final composed videos (MP4)
└── images/         # Downloaded Unsplash images
These directories are created automatically on backend startup by config.py:25-26.

​
Configuration File Setup

​
Step 1: Create .env File

Create a .env file in the backend/ directory: 
cd backend
cp .env.example .env

​
Step 2: Configure Variables

Edit the .env file with your API keys: 
# API Keys
GEMINI_API_KEY=your_gemini_api_key_here
SARVAM_API_KEY=your_sarvam_api_key_here
UNSPLASH_ACCESS_KEY=your_unsplash_access_key_here

# TTS Configuration (optional)
SARVAM_TTS_URL=https://api.sarvam.ai/text-to-speech
SARVAM_MODEL=bulbul:v2

# Server Configuration (optional)
HOST=0.0.0.0
PORT=8000
Never commit your .env file to version control. Keep your API keys secure and private.

​
Backend Configuration

The backend configuration is managed by config.py which loads environment variables and sets up paths:

​
Model Configuration

  • Gemini Model: gemini-2.5-flash (configured in config.py:29)
  • Manim Quality: m (medium quality, can be l/m/h)
  • Manim FPS: 30 frames per second

​
Language-Speaker Mapping

The system maps supported languages to Sarvam AI voice speakers:
LanguageSpeakerVoice Type
EnglishanushkaIndian English Female
HindimanishaHindi Female
KannadavidyaKannada Female
TeluguaryaTelugu Female
See Language Options for complete language support details.

​
Frontend Configuration

The frontend connects to the backend API server. Default configuration:
  • Frontend URL: http://localhost:5173
  • Backend URL: http://localhost:8000
  • CORS: Enabled for localhost development

​
CORS Settings

The backend allows CORS requests from:
  • http://localhost:5173
  • http://127.0.0.1:5173
To add additional origins, modify app.py:29.

​
System Requirements

​
Required Software

  1. Python 3.8+ - Backend runtime
  2. FFmpeg - Video processing and composition
  3. Manim - Mathematical animation engine
  4. Node.js 16+ - Frontend development server

​
Installing FFmpeg

  1. Download from ffmpeg.org
  2. Extract the archive
  3. Add ffmpeg/bin to System PATH
  4. Verify: ffmpeg -version

​
Installing Manim

pip install manim
Manim is used to generate mathematical animations and data visualizations for presentation slides.

​
Verification

After configuration, verify your setup:

​
Check Environment Variables

cd backend
python -c "from config import Config; print('API Keys configured:', bool(Config.GEMINI_API_KEY and Config.SARVAM_API_KEY and Config.UNSPLASH_ACCESS_KEY))"

​
Check Directory Structure

ls -la backend/outputs/
You should see all output directories listed.

​
Test Backend Server

cd backend
python app.py
Visit http://localhost:8000/health - you should see {"status": "healthy"}.

​
Troubleshooting

  • Verify .env file is in the backend/ directory
  • Check for typos in variable names
  • Ensure no extra spaces around the = sign
  • Restart the backend server after changes
  • Verify FFmpeg is installed: ffmpeg -version
  • On Windows, ensure FFmpeg is added to System PATH
  • Restart terminal after PATH changes
  • Ensure write permissions for backend/outputs/
  • On Linux/macOS: chmod -R 755 backend/outputs/
  • Check frontend is running on localhost:5173
  • Add your frontend URL to CORS settings in app.py
  • Clear browser cache and restart both servers

​
Next Steps

Configure API Keys

Learn how to obtain and configure API keys for all services

Language Options

Explore supported languages and voice customization

Build docs developers (and LLMs) love

Get started for freeTalk to us