Auto-Editor Documentation

repository·master·Indexed 26 days ago

https://github.com/wyattblue/auto-editor

A command-line application that automatically edits video and audio by analyzing audio loudness and motion to automate tasks like cutting out silence. It supports custom labels and actions for speed, volume, zoom, and rotation, as well as exporting edit data to Adobe Premiere Pro, DaVinci Resolve, Final Cut Pro, ShotCut, and Kdenlive.

Tokens
17.5K
Snippets
76
Records
113
Agent score
89%

What's inside Auto-Editor

  1. Use sidecar subtitles for the Transcribe → Cut workflow

    master

    If your media file does not have an embedded subtitle stream, auto-editor will automatically look for a sidecar file with the same name as the input but with a subtitle extension (e.g., talk.srt or talk.ass for talk.mp4).

    Workflow Example:

    1. Generate the subtitle file: auto-editor whisper talk.mp4 ggml-medium.en.bin --format srt -o talk.srt
    2. Edit the video using that file: auto-editor talk.mp4 --edit word:question
  2. Export projects to video editors

    master

    Instead of rendering a final video, you can export a project file for use in professional editing software using the --export flag.

    Supported formats include:

    • premiere
    • resolve
    • final-cut-pro
    • shotcut
    • kdenlive

    To export each kept section as an individual file, use --export clip-sequence.

  3. Deploy Auto-Editor using systemd

    master

    You can deploy the Auto-Editor website using a systemd service. Create a service file at /etc/systemd/system/ae.service with the following configuration. Note that ExecStart must point to your specific bun binary path and WorkingDirectory should match your deployment directory.

    [Unit]
    Description=AutoEditor Website
    After=network.target
    
    [Service]
    User=root
    WorkingDirectory=/var/www/auto-editor
    ExecStart=/root/.bun/bin/bun run server.js
    
    [Install]
    WantedBy=multi-user.target
  4. Cut or keep segments at the beginning or end of a video

    master

    To manually remove or preserve specific segments at the start or end of a video, use the start and end variables or negative indexes.

    • Cut the first and last 30 seconds: --cut start,30sec -30sec,end
    • Keep the first and last 30 seconds (overriding automatic edits): --keep start,30sec -30sec,end
    • Select the last 60 frames using negative indexes: -60,end
    • Select from the first second to the last 30 seconds: 1sec,-30secs
    auto-editor video.mp4 --cut start,30sec -30sec,end
  5. Normalize audio using EBU (R128) normalization

    master

    EBU (EBU R128) normalization analyzes perceived loudness over time to meet broadcast and streaming standards. It uses a two-pass process: an analysis pass to measure loudness, range, and true peak, followed by a normalization pass using the FFmpeg loudnorm filter.

    You can use default values, set a specific integrated loudness target, or provide a full string of parameters.

    auto-editor --audio-normalize ebu  # use default values
    auto-editor --audio-normalize ebu:i=-16  # set integrated loudness target
    auto-editor --audio-normalize "ebu:i=-5,lra=20,gain=5,tp=-1"  # customize all parameters
  6. Generate, render, and translate v3 timelines

    master

    The v3 format is a nonlinear timeline format (JSON-based with .v3 extension) that supports multiple overlapping video and audio layers. You can use Auto-Editor to:

    1. Generate a v3 timeline from media files using the --export v3 flag.
    2. Render media files from an existing v3 timeline.
    3. Translate other timeline formats (like Final Cut Pro 7 XML) into the v3 format.
  7. Install Auto-Editor via Official Binary (Recommended)

    master

    The recommended installation method is to use the official binaries for Windows, macOS, and Linux.

    1. Download the binary for your platform from the GitHub Releases page.
    2. Rename the downloaded file to auto-editor (or auto-editor.exe on Windows).
    3. On macOS or Linux, ensure the binary is executable by running chmod +x ./auto-editor.
    4. Move the binary to a directory in your PATH to make it accessible from any working directory.

    To verify the installation, run ./auto-editor --help.

    ./auto-editor --help
  8. Transcribe a live microphone

    master

    To transcribe live audio from a microphone, pass :mic as the file argument. This streams the microphone directly to the transcription engine without saving a media recording. To retain and edit the captured audio, use auto-editor :mic instead, which saves a sibling lossless-FLAC _RECORDING.mka file by default.

    Platform Support:

    • macOS: Uses AVFoundation (prefers USB, then system default).
    • Windows: Uses DirectShow (prefers USB, then first audio capture device).
    • Linux: Uses default ALSA input device.
  9. Basic usage of auto-editor

    master

    Use auto-editor to analyze a media file and automatically cut out inactive (e.g., silent) sections. By default, active sections are kept unchanged and inactive sections are cut.

    To run a basic edit, provide the input file path. To specify a custom output name, use the -o flag. Use --preview to see what would be cut without actually rendering the file.

    auto-editor path/to/video.mp4          # writes video_ALTERED.mp4
    auto-editor video.mp4 -o out.mp4       # choose the output name
    auto-editor --preview video.mp4        # sanity-check without rendering