Skip to main content

Subtitle Detect Language

FFmpeg Builder: Subtitle Detect Language

This flow element detects the language of subtitle tracks that do not already have a language code set, then updates the active FFmpeg Builder model.

It uses FileFlows Forge to identify text subtitle content. Image-based subtitle tracks are converted to text through Forge before detection.

note

This flow element requires FileFlows Forge to be configured and available.

How It Works​

  1. Iterates all non-deleted subtitle streams in the FFmpeg Builder model.
  2. Skips subtitle tracks that already have a known language code.
  3. Sends each subtitle stream to Forge for language identification.
  4. Converts image-based subtitle streams to text when needed.
  5. Updates the FFmpeg Builder subtitle language metadata with the detected ISO 639-2/T value.
  6. Marks the subtitle stream as changed so the FFmpeg Builder output includes the updated language metadata.

Language Handling​

Subtitle tracks are skipped when the language is already set to a known value.

Tracks are only checked when the language is blank, und, or unknown.

Detected languages are normalized to ISO 639-2/T style codes, for example:

  • eng for English
  • fra for French
  • spa for Spanish
  • deu for German
  • ita for Italian
  • jpn for Japanese

Outputs​

  1. One or more subtitle tracks had their language metadata detected and updated.
  2. No subtitle tracks needed language detection, or all detection attempts failed.