Play 94
AI Podcast Generator
Text-to-podcast pipeline with multi-speaker voice synthesis and music transitions.
Text-to-podcast pipeline with multi-speaker voice synthesis, music transitions, chapter markers, and content safety review. Converts blog posts, research papers, or meeting notes into professionally narrated audio content using Azure AI Voice Live SDK.
Architecture Pattern
Podcast pipeline: content ingestion - script generation - multi-voice synthesis - music layering - chapter marking - safety review
Azure Services
DevKit (.github Agentic OS)
- agent.md — root orchestrator with builder→reviewer→tuner handoffs
- 3 agents — Podcast Builder (gpt-4o), Reviewer (gpt-4o-mini), Tuner (gpt-4o-mini)
- 3 skills — deploy (217 lines), evaluate (105 lines), tune (228 lines)
- 4 prompts — /deploy, /test, /review, /evaluate with agent routing
- .vscode/mcp.json — FrootAI MCP with OpenAI + Speech inputs + envFile
TuneKit (AI Config)
- config/openai.json - script generation and content adaptation prompts
- config/podcast.json - voice personas, speaking rates, music styles
- config/guardrails.json - content safety thresholds, audio quality minimums
- evaluation/eval.py - Audio quality >4.0 MOS, Content fidelity >90%
Tuning Parameters
Machine evidence
FrootAI evidence lifecycle
This is an internal evidence maturity label, not third-party certification, accreditation, legal compliance, or a production guarantee. Missing or expired evidence demotes automatically; catalog claims cannot promote a play.
This play currently has design evidence only. A runnable scenario, endpoint evaluation, and build receipts are the next contiguous gates.
Repo Intelligence
v1A no-clone, revision-pinned map for agents and humans. Observed evidence is separated from inferred flow so the output stays useful without pretending to be a full call graph.