graph TD subgraph VTP[Video to Text Processing Agent] A[Extract MP3 from Video File] --> B[Process Audio via STT API] B --> C[Generate Raw Transcript] C --> D[Add Timestamp Markers] D --> E[Format to JSON Structure] E --> F[Validate JSON Output] end A1[Input: Video File] --> A F --> O1[Output: Timestamped JSON Transcript] style VTP fill:#f9f,stroke:#333,stroke-width:4px style A1 fill:#ccf,stroke:#333,stroke-width:2px style O1 fill:#cfc,stroke:#333,stroke-width:2px