Clips AI

Clips AI: AI Video Repurposing for Developers

What is Clips AI? Clips AI is an open-source Python library designed to automatically convert long-form videos into engaging clips. It's tailored for audio-centric and narrative-based content like podcasts, interviews, speeches, and sermons. With just a few lines of code, you can segment videos into multiple clips and resize their aspect ratio from 16:9 to 9:16.

Key Features:

• Automatic Clipping: Uses transcript analysis to identify key segments in the video.
• Intelligent Resizing: Dynamically reframes videos to focus on the current speaker, converting the video into various aspect ratios.

How to Use Clips AI:

1. Installation:

   • Install Python dependencies using pip:

     pip install clipsai
     pip install whisperx@git+https://github.com/m-bain/whisperx.git
     

   • Install libmagic and ffmpeg.

2. Creating Clips:

   • Transcribe the video using WhisperX (integrated in Clips AI).

   • Use ClipFinder to identify and create clips based on the transcription.

     from clipsai import ClipFinder, Transcriber
     
     transcriber = Transcriber()
     transcription = transcriber.transcribe(audio_file_path="/abs/path/to/video.mp4")
     
     clipfinder = ClipFinder()
     clips = clipfinder.find_clips(transcription=transcription)
     
     print("StartTime: ", clips[0].start_time)
     print("EndTime: ", clips[0].end_time)
     

3. Resizing a Video:

   • A Hugging Face access token is required for speaker diarization (using Pyannote).

     from clipsai import resize
     
     crops = resize(
         video_file_path="/abs/path/to/video.mp4",
         pyannote_auth_token="pyannote_token",
         aspect_ratio=(9, 16)
     )
     
     print("Crops: ", crops.segments)
     

Why is Clips AI Important?

Clips AI simplifies the process of repurposing long videos into short, engaging clips suitable for platforms like TikTok, Instagram Reels, and YouTube Shorts. It saves time and effort by automating the clipping and resizing process.

Where to Find More Information:

• GitHub: https://github.com/
• Demo:
• Quickstart: