Broadcast Captioning and Compliance Agent Deployment Plan
Deploy autonomous agents for real-time subtitling, audio description, and regulatory broadcast compliance.
Use this template when building and deploying autonomous speech and compliance agents for linear television or streaming networks. It establishes precise system instructions, latency guardrails, and automated failovers.
Role: Lead Broadcast Systems and AI Automation Architect
Context
- Network infrastructure: {{broadcast_network}}
- Target territories: {{target_territories}}
- Content asset type: {{asset_catalog_type}}
- Regulatory standards: {{compliance_standards}}
- Real-time latency budget: {{latency_threshold}}
- Failure handling: {{fallback_protocol}}
Task
Build a complete operational plan and system prompt framework for autonomous agents generating real-time closed captions, audio descriptions, and compliance flags for live broadcast media within {{broadcast_network}}.
Method
- Map regulatory requirements and terminology dictionaries across {{target_territories}} based on {{compliance_standards}}.
- Define audio ingestion, speech-to-text diarization, and context parsing pipelines within the latency envelope of {{latency_threshold}}.
- Formulate the agent system prompt architecture covering caption timing, speaker identification, and tone modulation for {{asset_catalog_type}}.
- Integrate real-time content filtering guardrails to flag profanity, hate speech, and sensitive geopolitical terms across {{target_territories}}.
- Establish automated quality scoring routines measuring Word Error Rate (WER) and timing synchronization drift.
- Detail failover triggers that shift traffic to {{fallback_protocol}} upon consecutive latency spikes or hallucination flags.
- Specify human-in-the-loop escalation workflows for live override consoles at {{broadcast_network}}.
Constraints
- MUST ensure caption latency does not exceed {{latency_threshold}} under peak live streaming load.
- MUST NOT emit unverified compliance clearance signals without explicit confidence threshold validation.
- Include explicit domain glossaries for high-frequency broadcast jargon.
- Prioritize zero-downtime cutover strategies across all distribution endpoints.
Output format
- Executive Architecture Summary (150-200 words)
- Core Agent Instruction Set & System Prompt (Complete, copyable prompt block)
- Ingestion & Latency Management Matrix (Table of 4-6 pipeline stages)
- Compliance & Guardrail Protocol (Structured checklist)
- Failover & Override Procedures (Step-by-step workflow)
Self-review
- Verify all 6 variables are referenced naturally and accurately.
- Confirm system prompt includes concrete speech-to-text formatting directives.
- Ensure compliance checks align directly with {{compliance_standards}}.
- Check that total character count fits inside the target range.
Explicit role, a named task, and discrete steps the model can follow.
Background, inputs and variables the model needs before it starts.
Hard boundaries — what the model must and must not do.
A named, field-level shape for the response.
Ordered work items that force analysis before an answer.
Length and structure that travel across frontier models.
Signal density — instruction weight without padding.
Documented variables so the scaffold adapts to new inputs.
Quality bar, assumptions and behaviour when inputs are thin.
How much real usage the template has behind it.