Workflow nodes supported by the Media AI solution and their capabilities.
|
Node Name |
Description |
|
Face Recognition |
Takes multiple actor images, names, and roles as input. Detects faces in video and identifies when each character appears. |
|
Speech Extraction + ASR |
Extracts speech from audio/video and runs high-accuracy ASR to produce speaker-attributed, timestamped transcripts. |
|
Text Content Extraction |
Select AI models, configure a prompt, and define the output format. Combines ASR transcripts with frame-sampled visual analysis for cross-modal text extraction and comprehensive semantic understanding. |
|
Video Summary (OTT) |
Generates summaries and recaps for various video formats (e.g., movies, TV series, and short videos). Analyzes visual frames at configurable intervals and similarity thresholds, combined with subtitle and ASR content. |
|
Frame Extraction & Analysis (Custom) |
Samples frames at configurable intervals and similarity thresholds. Select a model and define a custom prompt and output format to analyze frames for content, quality, or other visual attributes. |