Create Educational Videos with Python | Generated by AI

Home PDF

To create educational short videos with a black background and captions synchronized with audio generated from text using the DeepSeek API and Google Cloud Text-to-Speech API, you can use Python to orchestrate the process. Below is a step-by-step guide and a Python script that accomplishes this. The script will:

  1. Use DeepSeek API to generate or refine a script (assuming you provide the educational content).
  2. Use Google Cloud Text-to-Speech API to convert the script into audio.
  3. Use a library like moviepy to create a video with a black background and captions synchronized with the audio.

Prerequisites

Steps

  1. Generate or Refine Script with DeepSeek API: Use DeepSeek to create or polish the educational script, ensuring it’s concise and suitable for a 1-minute video.
  2. Convert Text to Audio with Google Cloud Text-to-Speech: Split the script into paragraphs, generate audio for each, and save as separate audio files.
  3. Create Video with MoviePy: Generate a video with a black background, display captions for each paragraph synchronized with the audio, and combine them into a final 1-minute video.

Python Script

The following script assumes you have a text file with the educational content (paragraphs) and generates a video with a black background and captions.

import os
from openai import OpenAI
from google.cloud import texttospeech
from moviepy.editor import TextClip, CompositeVideoClip, ColorClip, concatenate_videoclips
import requests

# Set up environment variables
os.environ["GOOGLE_APPLICATION_CREDENTIALS"] = "path/to/your/google-credentials.json"  # Update with your credentials file path
DEEPSEEK_API_KEY = "your_deepseek_api_key"  # Update with your DeepSeek API key

# Initialize DeepSeek client
deepseek_client = OpenAI(api_key=DEEPSEEK_API_KEY, base_url="https://api.deepseek.com")

# Function to refine script with DeepSeek
def refine_script_with_deepseek(script):
    prompt = f"""
    You are an expert scriptwriter for educational videos. Refine the following script to be concise, clear, and engaging for a 1-minute educational video. Ensure it is suitable for spoken narration and fits within 60 seconds when spoken at a natural pace. Split the script into 2-3 short paragraphs for caption display. Return the refined script as a list of paragraphs.

    Original script:
    {script}

    Output format:
    ["paragraph 1", "paragraph 2", "paragraph 3"]
    """
    response = deepseek_client.chat.completions.create(
        model="deepseek-chat",
        messages=[{"role": "user", "content": prompt}],
        stream=False
    )
    refined_script = eval(response.choices[0].message.content)  # Convert string to list
    return refined_script

# Function to generate audio for each paragraph using Google Cloud Text-to-Speech
def generate_audio(paragraphs, output_dir="audio"):
    client = texttospeech.TextToSpeechClient()
    if not os.path.exists(output_dir):
        os.makedirs(output_dir)
    
    audio_files = []
    for i, paragraph in enumerate(paragraphs):
        synthesis_input = texttospeech.SynthesisInput(text=paragraph)
        voice = texttospeech.VoiceSelectionParams(
            language_code="en-US",
            name="en-US-Wavenet-D"  # A natural-sounding English voice
        )
        audio_config = texttospeech.AudioConfig(
            audio_encoding=texttospeech.AudioEncoding.MP3
        )
        response = client.synthesize_speech(
            input=synthesis_input, voice=voice, audio_config=audio_config
        )
        audio_file = os.path.join(output_dir, f"paragraph_{i+1}.mp3")
        with open(audio_file, "wb") as out:
            out.write(response.audio_content)
        audio_files.append(audio_file)
    return audio_files

# Function to create video with captions and black background
def create_video(paragraphs, audio_files, output_file="educational_video.mp4"):
    clips = []
    for i, (paragraph, audio_file) in enumerate(zip(paragraphs, audio_files)):
        # Create text clip for captions
        text_clip = TextClip(
            paragraph,
            fontsize=40,
            color="white",
            font="Arial",
            size=(1280, 720),  # Standard HD resolution
            method="caption",
            align="center"
        )
        # Load audio and get its duration
        audio_clip = AudioFileClip(audio_file)
        duration = audio_clip.duration
        # Set text clip duration to match audio
        text_clip = text_clip.set_duration(duration)
        # Create black background clip
        bg_clip = ColorClip(size=(1280, 720), color=(0, 0, 0), duration=duration)
        # Combine text and background
        video_clip = CompositeVideoClip([bg_clip, text_clip.set_position("center")])
        # Add audio to video clip
        video_clip = video_clip.set_audio(audio_clip)
        clips.append(video_clip)
    
    # Concatinate all clips
    final_clip = concatenate_videoclips(clips)
    # Write final video
    final_clip.write_videofile(output_file, fps=24, codec="libx264", audio_codec="aac")
    final_clip.close()
    for clip in clips:
        clip.close()

# Main function
def main():
    # Example input script (replace with your educational content)
    input_script = """
    Machine learning is a field of artificial intelligence that allows computers to learn from data without being explicitly programmed. It involves algorithms that identify patterns and make predictions. Applications include image recognition, natural language processing, and more. This technology is transforming industries like healthcare and finance.
    """
    
    # Step 1: Refine script with DeepSeek
    refined_paragraphs = refine_script_with_deepseek(input_script)
    print("Refined Script:", refined_paragraphs)
    
    # Step 2: Generate audio for each paragraph
    audio_files = generate_audio(refined_paragraphs)
    print("Audio files generated:", audio_files)
    
    # Step 3: Create video with captions and black background
    create_video(refined_paragraphs, audio_files)
    print("Video created: educational_video.mp4")

if __name__ == "__main__":
    main()

How to Use

  1. Set Up Credentials:
    • Replace "path/to/your/google-credentials.json" with the path to your Google Cloud service account JSON file.
    • Replace "your_deepseek_api_key" with your DeepSeek API key.
  2. Prepare Input Script:
    • Modify the input_script variable in the main() function with your educational content. The script should be a single string with the full text you want to convert into a video.
  3. Run the Script:
    • Save the script as create_educational_video.py and run it with python create_educational_video.py.
    • The script will:
      • Refine the script using DeepSeek API to ensure it’s concise and split into 2-3 paragraphs.
      • Generate MP3 audio files for each paragraph using Google Cloud Text-to-Speech.
      • Create a video with a black background, displaying each paragraph as captions synchronized with its corresponding audio.
  4. Output:
    • The final video will be saved as educational_video.mp4 in the same directory as the script.
    • Audio files for each paragraph will be saved in the audio directory.

Notes

Example Output

If your input script is about machine learning, the refined script might look like:

["Machine learning, a branch of AI, enables computers to learn from data.", 
 "It uses algorithms to find patterns and predict outcomes.", 
 "Applications include image recognition and healthcare innovations."]

This approach is simple, cost-effective, and produces professional-looking educational videos suitable for platforms like YouTube or educational websites. If you need further customization (e.g., different fonts, caption styles, or additional effects), let me know!

Back 2025.06.29 Donate