Debuly

Clips AI

Open-source Python library that finds clips in long videos from the transcript and reframes them to vertical.

Open source

What is Clips AI?

Clips AI is an open-source Python library for developers who want to convert long-form video into clips in code. It transcribes a video with WhisperX, then finds clip boundaries from the transcript using a topic-shift method (TextTiling with BERT embeddings) that suits audio-driven, narrative content such as podcasts, interviews, speeches and sermons. A separate resizing function reframes the video to another aspect ratio, 9:16 by default, keeping the current speaker in frame by combining speaker diarization with Pyannote, scene change detection with PySceneDetect and face detection with MTCNN and MediaPipe. A media editor then trims or resizes the files. Install it with pip, along with WhisperX, libmagic and ffmpeg. The documentation lists no paid plans.

Use cases

  • Split podcasts into short clips
  • Find clip boundaries from a transcript
  • Reframe 16:9 video to 9:16
  • Keep the active speaker in frame
  • Transcribe audio with word timestamps
  • Build a clipping pipeline in Python

Who it's for

  • Developers
  • Video tool builders
  • Podcast producers

Why use Clips AI

  • It runs in your own Python code, so you keep control of the pipeline.
  • Installation uses pip with open-source building blocks.
  • Clipping and reframing are separate steps, so you can use either one alone.
  • Speaker-aware cropping keeps the person talking in a vertical frame.
  • Language can be detected automatically or set explicitly.

Clips AI features

  • Transcript-based clip finding using TextTiling with BERT embeddings
  • Transcription through WhisperX with word-level timestamps
  • Automatic language detection or a chosen ISO 639-1 language code
  • Speaker-focused resizing to 9:16 or other aspect ratios
  • Speaker diarization with Pyannote
  • Scene change detection with PySceneDetect
  • Face detection with MTCNN and MediaPipe
  • Media editor for trimming and resizing files

Clips AI integrations

  • WhisperX
  • Pyannote
  • PySceneDetect
  • MediaPipe
  • FFmpeg

How to get started with Clips AI

  1. Install the library with pip, then install WhisperX, libmagic and ffmpeg.
  2. Transcribe the video with the Transcriber class to get word-level timestamps.
  3. Run ClipFinder on the transcription to get start and end times for clips.
  4. Call resize with a Pyannote token and aspect ratio, then use the media editor to trim and resize.

Clips AI limitations

  • Designed for audio-centric, narrative videos such as podcasts, interviews, speeches and sermons.
  • Resizing needs a Pyannote access token from Hugging Face.
  • Requires installing WhisperX, libmagic and ffmpeg, and you run it on your own hardware.
  • The documentation does not list a hosted service.

Clips AI pricing

Clips AI is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

Clips AI alternatives

The main Clips AI alternatives listed on Debuly are OpusClip, Descript, WayinVideo, ClipSpeedAI, Hipclip and Veo. The table compares pricing and platforms; order is A–Z.

Clips AI compared with its alternatives
ToolPricingStarting pricePlatforms
Clips AIOpen source—Python library, Self-hosted
ClipSpeedAIFree trial$19/moWeb, API
DescriptFreemium$16/moWeb
HipclipFree trial$20/moWeb
OpusClipFreemium$15/moWeb
VeoPricing unknown—Web, API
WayinVideoFreemium$9.58/moWeb, API
  1. ClipSpeedAI

    Turns long videos and podcasts into captioned vertical clips, with scheduling, an API, and AI video and image generation.

  2. Descript

    Video and podcast editor where you cut footage by editing its transcript, with AI for audio cleanup, clips, dubbing and avatars.

  3. Hipclip

    AI co-editor that turns long videos and podcasts into short clips, captions, posts and newsletters.

  4. OpusClip

    AI video clipper that cuts long videos into captioned, reframed short clips and posts them to social platforms.

  5. Veo

    Google DeepMind's video generation model that turns text prompts or images into clips with synchronized sound.

  6. WayinVideo

    AI video platform that clips long videos into shorts, finds moments by text search, and adds captions and summaries.

Clips AI FAQ

Is Clips AI open source?

Yes. The documentation describes it as an open-source Python library you install with pip.

How much does Clips AI cost?

The documentation lists no paid plans or prices. You do need the compute to run transcription and resizing, plus a Pyannote token from Hugging Face.

What kind of videos does Clips AI work best with?

It is designed for audio-centric, narrative videos such as podcasts, interviews, speeches and sermons, because clips are found from the transcript.

How does Clips AI reframe video for vertical formats?

It uses speaker diarization, scene change detection and face detection to crop the video around whoever is speaking, with 9:16 as the default ratio.

Does Clips AI have an API?

It is a Python library with classes such as Transcriber and ClipFinder and a resize function, rather than a hosted web API.

What are alternatives to Clips AI?

Hosted alternatives for AI clipping include OpusClip, Vizard and Descript.

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

SourceTypeAccessed
clipsai.com/official
clipsai.com/official
clipsai.com/references/clipofficial
clipsai.com/references/resizeofficial
clipsai.com/references/transcribeofficial

Image credits

  • Logo: none collected yet, so a monogram is shown.
  • Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.