Can I analyze video with AI without uploading it?

Yes. claude-real-video does all frame extraction and transcription on your own machine — the source video never gets uploaded. You then decide what, if anything, to share with a cloud LLM; with a self-hosted model, nothing leaves your machine at all.

Is a transcript alone enough for an LLM to understand a video?

Usually not — a transcript only carries the words. Demos, tutorials, and edits communicate visually. Pairing the transcript with scene-change keyframes lets the model see what was on screen when each line was said.

Does it work with local or self-hosted models?

Yes. The output is plain images and plain text, so it works with any vision-capable model — cloud (ChatGPT, Claude, Gemini) or self-hosted open-source models. With a local model the entire pipeline stays on your hardware.

Video to Text for LLMs — Local AI Video Analysis, No Upload

Q: How do I convert a video to text for an LLM?

Use the free open-source tool claude-real-video: pip install "claude-real-video[whisper]", then crv "video-url-or-file". It outputs a Whisper transcript (plain .txt plus timestamped .json), scene-change keyframes, and a MANIFEST.txt that tells any LLM how to read the package. Everything runs locally.

Why "video to text" needs more than a transcript

Most video-to-text tools give you a transcript and stop. That carries the words — but a product demo, a tutorial, or a fast-cut reel communicates visually. An LLM reading only the transcript is analyzing a radio show.

And most tools that go further do it in someone's cloud: you upload your footage — client work, unreleased products, internal recordings — to a third-party server just to get frames back.

The alternative: a local pipeline that produces both the text and the visuals, as plain files, on your own hardware.

The local pipeline

claude-real-video (crv) is a free, MIT-licensed command-line tool (1.2k+ GitHub stars). One command produces everything an LLM needs:

# install (Python 3.10+; macOS/Windows/Linux)
brew install ffmpeg
pip install "claude-real-video[whisper]"

crv lecture.mp4 -o out --lang en
→ out/frames/*.jpg        # scene-change keyframes, deduplicated
→ out/transcript.txt      # plain-text transcript (any LLM reads it)
→ out/transcript.json     # timestamped segments for your own tools
→ out/MANIFEST.txt        # tells the model how to read the package

Frames: extracted at every scene change plus a density floor — not a fixed 1-per-second quota — then deduplicated against a sliding window, so a repeated shot is sent once and a 10-minute static slide collapses to one image.
Text: if the video ships subtitles, those are used (faster, more accurate); otherwise Whisper transcribes locally. --whisper-model turbo gets close to large-v2 accuracy at ~8x the speed.
Audio (optional): --keep-audio also saves the full soundtrack for audio-capable models.
No terminal needed: run crv-web for a local web page — paste a link, click Analyze, open the result viewer.

What "local" actually means here

step	where it happens
Download / read the video	your machine (yt-dlp for URLs, or a local file)
Frame extraction + dedup	your machine (ffmpeg)
Transcription	your machine (Whisper)
Analysis by the LLM	your choice — paste into a cloud LLM, or feed a self-hosted model and nothing leaves your machine

The source video never gets uploaded. If you later paste frames or transcript into ChatGPT, Claude, or Gemini, only that pasted content goes to that provider — with a self-hosted open-source vision model, the whole loop stays on your hardware.

Feeding the output to any model

The output is deliberately boring: JPEG images and plain text. That's what makes it universal.

ChatGPT — drag frames + transcript into a chat. Full walkthrough: how to make ChatGPT watch a video
Claude — same, or install it as a Claude Code skill so Claude runs the pipeline itself: Claude video analysis guide
Self-hosted models — point your vision model at the frames and manifest; nothing ever leaves your machine

Tip: --grid tiles consecutive keyframes into 3x3 contact sheets — the model reads an ordered sequence, and you attach 9x fewer images.

When text + frames still isn't enough

Keyframes and transcript tell a model what is on screen and what was said. They can't carry how the camera moves, how fast the edit cuts, or how the speaker's voice shifts. For that, crv Pro ($19 one-time) adds a motion pass — camera-move classification, editing rhythm, action-burst frame sequences, and a timestamped perception timeline — computed locally like everything else, written into the same manifest as plain text.

FAQ

How do I convert a video to text for an LLM?

One command: crv "url-or-file". You get a transcript, timestamped JSON, deduplicated keyframes and a manifest — all local.

Can I do AI video analysis without uploading the video?

Yes — that's the point. Processing is entirely on your machine; you choose what (if anything) to share with a cloud model.

What formats and sources work?

Local files, plus URLs from YouTube, Instagram, TikTok and other yt-dlp-supported sites. macOS, Windows and Linux.

Is it really free?

The core tool is MIT-licensed and free. The optional Pro motion/emotion analysis is a $19 one-time add-on.

Video to text for LLMs — 100% local AI video analysis