graph TD A[Start] --> B[Extract MP3 from Video] B --> C[Initialize Speech-to-Text API] C --> D[Process Audio Through API] D --> E[Generate Raw Transcript] E --> F[Add Timestamp Markers] F --> G[Format as JSON] G --> H[Validate JSON Structure] H --> I[Error Check] I -->|No Errors| J[Save Final JSON] I -->|Errors Found| F J --> K[End] subgraph Transcription Processor B C D E F G H I J end