@twick/ai-models / Exports / CaptionGenerationInput
Interface: CaptionGenerationInput
Table of contents
Properties
Properties
audioUrl
• Optional audioUrl: string
Defined in
autoDetectLanguage
• Optional autoDetectLanguage: boolean
When true (default for ai-caption jobs), the backend should attempt to auto-detect the spoken language from the audio instead of relying solely on the explicit language field.
Defined in
language
• Optional language: string
Defined in
languageFont
• Optional languageFont: string
Defined in
phraseLength
• Optional phraseLength: CaptionPhraseLength
Desired subtitle phrase granularity. The backend maps this preset to pause and duration-aware segmentation thresholds instead of fixed word chunking.
Defined in
videoUrl
• Optional videoUrl: string
Defined in
wordsPerPhrase
• Optional wordsPerPhrase: number
Deprecated
Use phraseLength instead.