Interface VideoAnalyzeRequestRich

Rich video analysis request with scene detection, transcription, and RAG indexing support.

Extends the simpler VideoAnalyzeRequest pattern with fine-grained control over scene detection thresholds, description detail, and optional RAG indexing of analysis results.

Example

const request: VideoAnalyzeRequestRich = {
  video: 'https://example.com/demo.mp4',
  sceneThreshold: 0.3,
  transcribeAudio: true,
  descriptionDetail: 'detailed',
  onProgress: (evt) => console.log(`${evt.phase}: ${evt.progress}%`),
};
interface VideoAnalyzeRequestRich {
    video: string | Buffer<ArrayBufferLike>;
    prompt?: string;
    sceneThreshold?: number;
    transcribeAudio?: boolean;
    descriptionDetail?: DescriptionDetail;
    maxFrames?: number;
    maxScenes?: number;
    indexForRAG?: boolean;
    onProgress?: ((event) => void);
}

Properties

video prompt? sceneThreshold? transcribeAudio? descriptionDetail? maxFrames? maxScenes? indexForRAG? onProgress?

Properties

video

video: string | Buffer<ArrayBufferLike>

Video to analyze — either a URL string or a raw Buffer. When a URL is provided, the pipeline downloads the video to a temporary file before processing.

Optional prompt

prompt?: string

Optional analysis prompt or question that should guide the final answer. When omitted, the analyzer produces a general-purpose summary.

Optional sceneThreshold

sceneThreshold?: number

Threshold for scene change detection (0-1). Lower values detect more scene boundaries (more sensitive); higher values only detect dramatic cuts.

Default

0.3

Optional transcribeAudio

transcribeAudio?: boolean

Whether to transcribe the audio track using Whisper. When enabled, each scene's transcript is populated and a full transcript is included in the analysis.

Default

true

Optional descriptionDetail

descriptionDetail?: DescriptionDetail

How detailed scene descriptions should be.

Default

'detailed'

Optional maxFrames

maxFrames?: number

Maximum number of frames to sample from the extracted frame set. When the extracted frame count exceeds this value, frames are evenly downsampled before scene detection and description.

Optional maxScenes

maxScenes?: number

Maximum number of scenes to detect. Prevents runaway analysis on very long videos with many cuts.

Default

100

Optional indexForRAG

indexForRAG?: boolean

Whether to index scene descriptions and transcripts into the RAG vector store for later retrieval.

Default

false

Optional onProgress

onProgress?: ((event) => void)

Optional callback invoked as analysis progresses through phases. Called with a VideoAnalysisProgressEvent at each phase transition and when per-scene progress updates are available.

Type declaration

Settings

Member Visibility

Theme

On This Page

videopromptsceneThresholdtranscribeAudiodescriptionDetailmaxFramesmaxScenesindexForRAGonProgress