[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7DHKE3RF2DB2AFH2JV4NV8Y":3,"skill-same:01M7DHKE3RF2DB2AFH2JV4NV8Y":60},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":23,"commitSha":24,"skillDirectory":25,"forksCount":13,"repositoryHtmlUrl":26,"repositoryHomepageUrl":27,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":29,"install":30,"resources":43},"01M7DHKE3RF2DB2AFH2JV4NV8Y","podcast-generation",null,"Generate AI-powered podcast-style audio narratives using Azure OpenAI's GPT Realtime Mini model via WebSocket. Use when building text-to-speech features, audio narrative generation, podcast creation from content, or integrating with Azure OpenAI Realtime API for real audio output. Covers full-stack implementation from React frontend to Python FastAPI backend with WebSocket streaming.","ak2dev\u002Fagentguard","ak2dev","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F233005497?v=4","01M7DGWK3E3XCC276C6YRFSN05","01M7DGWK4J2A9WQ44CS5CRP4R6",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fak2dev\u002Fagentguard\u002Fblob\u002Fmain\u002Fbench\u002Fcorpus\u002Fbenign\u002Freal-ms-podcast-generation\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: podcast-generation\ndescription: Generate AI-powered podcast-style audio narratives using Azure OpenAI's GPT Realtime Mini model via WebSocket. Use when building text-to-speech features, audio narrative generation, podcast creation from content, or integrating with Azure OpenAI Realtime API for real audio output. Covers full-stack implementation from React frontend to Python FastAPI backend with WebSocket streaming.\n---\n\n# Podcast Generation with GPT Realtime Mini\n\nGenerate real audio narratives from text content using Azure OpenAI's Realtime API.\n\n## Quick Start\n\n1. Configure environment variables for Realtime API\n2. Connect via WebSocket to Azure OpenAI Realtime endpoint\n3. Send text prompt, collect PCM audio chunks + transcript\n4. Convert PCM to WAV format\n5. Return base64-encoded audio to frontend for playback\n\n## Environment Configuration\n\n```env\nAZURE_OPENAI_AUDIO_API_KEY=your_realtime_api_key\nAZURE_OPENAI_AUDIO_ENDPOINT=https:\u002F\u002Fyour-resource.cognitiveservices.azure.com\nAZURE_OPENAI_AUDIO_DEPLOYMENT=gpt-realtime-mini\n```\n\n**Note**: Endpoint should NOT include `\u002Fopenai\u002Fv1\u002F` - just the base URL.\n\n## Core Workflow\n\n### Backend Audio Generation\n\n```python\nfrom openai import AsyncOpenAI\nimport base64\n\n# Convert HTTPS endpoint to WebSocket URL\nws_url = endpoint.replace(\"https:\u002F\u002F\", \"wss:\u002F\u002F\") + \"\u002Fopenai\u002Fv1\"\n\nclient = AsyncOpenAI(\n    websocket_base_url=ws_url,\n    api_key=api_key\n)\n\naudio_chunks = []\ntranscript_parts = []\n\nasync with client.realtime.connect(model=\"gpt-realtime-mini\") as conn:\n    # Configure for audio-only output\n    await conn.session.update(session={\n        \"output_modalities\": [\"audio\"],\n        \"instructions\": \"You are a narrator. Speak naturally.\"\n    })\n    \n    # Send text to narrate\n    await conn.conversation.item.create(item={\n        \"type\": \"message\",\n        \"role\": \"user\",\n        \"content\": [{\"type\": \"input_text\", \"text\": prompt}]\n    })\n    \n    await conn.response.create()\n    \n    # Collect streaming events\n    async for event in conn:\n        if event.type == \"response.output_audio.delta\":\n            audio_chunks.append(base64.b64decode(event.delta))\n        elif event.type == \"response.output_audio_transcript.delta\":\n            transcript_parts.append(event.delta)\n        elif event.type == \"response.done\":\n            break\n\n# Convert PCM to WAV (see scripts\u002Fpcm_to_wav.py)\npcm_audio = b''.join(audio_chunks)\nwav_audio = pcm_to_wav(pcm_audio, sample_rate=24000)\n```\n\n### Frontend Audio Playback\n\n```javascript\n\u002F\u002F Convert base64 WAV to playable blob\nconst base64ToBlob = (base64, mimeType) => {\n  const bytes = atob(base64);\n  const arr = new Uint8Array(bytes.length);\n  for (let i = 0; i \u003C bytes.length; i++) arr[i] = bytes.charCodeAt(i);\n  return new Blob([arr], { type: mimeType });\n};\n\nconst audioBlob = base64ToBlob(response.audio_data, 'audio\u002Fwav');\nconst audioUrl = URL.createObjectURL(audioBlob);\nnew Audio(audioUrl).play();\n```\n\n## Voice Options\n\n| Voice | Character |\n|-------|-----------|\n| alloy | Neutral |\n| echo | Warm |\n| fable | Expressive |\n| onyx | Deep |\n| nova | Friendly |\n| shimmer | Clear |\n\n## Realtime API Events\n\n- `response.output_audio.delta` - Base64 audio chunk\n- `response.output_audio_transcript.delta` - Transcript text\n- `response.done` - Generation complete\n- `error` - Handle with `event.error.message`\n\n## Audio Format\n\n- **Input**: Text prompt\n- **Output**: PCM audio (24kHz, 16-bit, mono)\n- **Storage**: Base64-encoded WAV\n\n## References\n\n- **Full architecture**: See [references\u002Farchitecture.md](references\u002Farchitecture.md) for complete stack design\n- **Code examples**: See [references\u002Fcode-examples.md](references\u002Fcode-examples.md) for production patterns\n- **PCM conversion**: Use [scripts\u002Fpcm_to_wav.py](scripts\u002Fpcm_to_wav.py) for audio format conversion\n","# Podcast Generation with GPT Realtime Mini\n\nGenerate real audio narratives from text content using Azure OpenAI's Realtime API.\n\n## Quick Start\n\n1. Configure environment variables for Realtime API\n2. Connect via WebSocket to Azure OpenAI Realtime endpoint\n3. Send text prompt, collect PCM audio chunks + transcript\n4. Convert PCM to WAV format\n5. Return base64-encoded audio to frontend for playback\n\n## Environment Configuration\n\n```env\nAZURE_OPENAI_AUDIO_API_KEY=your_realtime_api_key\nAZURE_OPENAI_AUDIO_ENDPOINT=https:\u002F\u002Fyour-resource.cognitiveservices.azure.com\nAZURE_OPENAI_AUDIO_DEPLOYMENT=gpt-realtime-mini\n```\n\n**Note**: Endpoint should NOT include `\u002Fopenai\u002Fv1\u002F` - just the base URL.\n\n## Core Workflow\n\n### Backend Audio Generation\n\n```python\nfrom openai import AsyncOpenAI\nimport base64\n\n# Convert HTTPS endpoint to WebSocket URL\nws_url = endpoint.replace(\"https:\u002F\u002F\", \"wss:\u002F\u002F\") + \"\u002Fopenai\u002Fv1\"\n\nclient = AsyncOpenAI(\n    websocket_base_url=ws_url,\n    api_key=api_key\n)\n\naudio_chunks = []\ntranscript_parts = []\n\nasync with client.realtime.connect(model=\"gpt-realtime-mini\") as conn:\n    # Configure for audio-only output\n    await conn.session.update(session={\n        \"output_modalities\": [\"audio\"],\n        \"instructions\": \"You are a narrator. Speak naturally.\"\n    })\n    \n    # Send text to narrate\n    await conn.conversation.item.create(item={\n        \"type\": \"message\",\n        \"role\": \"user\",\n        \"content\": [{\"type\": \"input_text\", \"text\": prompt}]\n    })\n    \n    await conn.response.create()\n    \n    # Collect streaming events\n    async for event in conn:\n        if event.type == \"response.output_audio.delta\":\n            audio_chunks.append(base64.b64decode(event.delta))\n        elif event.type == \"response.output_audio_transcript.delta\":\n            transcript_parts.append(event.delta)\n        elif event.type == \"response.done\":\n            break\n\n# Convert PCM to WAV (see scripts\u002Fpcm_to_wav.py)\npcm_audio = b''.join(audio_chunks)\nwav_audio = pcm_to_wav(pcm_audio, sample_rate=24000)\n```\n\n### Frontend Audio Playback\n\n```javascript\n\u002F\u002F Convert base64 WAV to playable blob\nconst base64ToBlob = (base64, mimeType) => {\n  const bytes = atob(base64);\n  const arr = new Uint8Array(bytes.length);\n  for (let i = 0; i \u003C bytes.length; i++) arr[i] = bytes.charCodeAt(i);\n  return new Blob([arr], { type: mimeType });\n};\n\nconst audioBlob = base64ToBlob(response.audio_data, 'audio\u002Fwav');\nconst audioUrl = URL.createObjectURL(audioBlob);\nnew Audio(audioUrl).play();\n```\n\n## Voice Options\n\n| Voice | Character |\n|-------|-----------|\n| alloy | Neutral |\n| echo | Warm |\n| fable | Expressive |\n| onyx | Deep |\n| nova | Friendly |\n| shimmer | Clear |\n\n## Realtime API Events\n\n- `response.output_audio.delta` - Base64 audio chunk\n- `response.output_audio_transcript.delta` - Transcript text\n- `response.done` - Generation complete\n- `error` - Handle with `event.error.message`\n\n## Audio Format\n\n- **Input**: Text prompt\n- **Output**: PCM audio (24kHz, 16-bit, mono)\n- **Storage**: Base64-encoded WAV\n\n## References\n\n- **Full architecture**: See [references\u002Farchitecture.md](references\u002Farchitecture.md) for complete stack design\n- **Code examples**: See [references\u002Fcode-examples.md](references\u002Fcode-examples.md) for production patterns\n- **PCM conversion**: Use [scripts\u002Fpcm_to_wav.py](scripts\u002Fpcm_to_wav.py) for audio format conversion\n",{"name":5,"description":7},"main","634c25395a24b98cd6125453ad4abb65517bcabb","bench\u002Fcorpus\u002Fbenign\u002Freal-ms-podcast-generation","https:\u002F\u002Fgithub.com\u002Fak2dev\u002Fagentguard","https:\u002F\u002Fsite-bionic8.vercel.app","https:\u002F\u002Fgithub.com\u002Fak2dev","2026-11-11 02:37:06",{"source":8,"skillName":5,"commands":31,"prompt":41,"zipAvailable":42},[32,35,38],{"tool":33,"command":34},"npx","npx skills add ak2dev\u002Fagentguard --skill podcast-generation",{"tool":36,"command":37},"bunx","bunx skills add ak2dev\u002Fagentguard --skill podcast-generation",{"tool":39,"command":40},"pnpm","pnpm dlx skills add ak2dev\u002Fagentguard --skill podcast-generation","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fak2dev\u002Fagentguard\u002Fblob\u002Fmain\u002Fbench\u002Fcorpus\u002Fbenign\u002Freal-ms-podcast-generation\u002FSKILL.md",true,[44,50,54],{"path":45,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":48,"sourceHtmlUrl":49},"references\u002Farchitecture.md","REFERENCE","参考文档",4736,"https:\u002F\u002Fgithub.com\u002Fak2dev\u002Fagentguard\u002Fblob\u002Fmain\u002Fbench\u002Fcorpus\u002Fbenign\u002Freal-ms-podcast-generation\u002Freferences\u002Farchitecture.md",{"path":51,"resourceType":46,"resourceTypeDesc":47,"sizeBytes":52,"sourceHtmlUrl":53},"references\u002Fcode-examples.md",6116,"https:\u002F\u002Fgithub.com\u002Fak2dev\u002Fagentguard\u002Fblob\u002Fmain\u002Fbench\u002Fcorpus\u002Fbenign\u002Freal-ms-podcast-generation\u002Freferences\u002Fcode-examples.md",{"path":55,"resourceType":56,"resourceTypeDesc":57,"sizeBytes":58,"sourceHtmlUrl":59},"scripts\u002Fpcm_to_wav.py","SCRIPT","脚本",2475,"https:\u002F\u002Fgithub.com\u002Fak2dev\u002Fagentguard\u002Fblob\u002Fmain\u002Fbench\u002Fcorpus\u002Fbenign\u002Freal-ms-podcast-generation\u002Fscripts\u002Fpcm_to_wav.py",[61,68,73,78,83,89],{"skillNo":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":18},"01M7DJ7GBSJBJ32GXNQXVAMNYK","csv-tool","Summarize CSV files and report column statistics.","https:\u002F\u002Fgithub.com\u002Fak2dev\u002Fagentguard\u002Fblob\u002Fmain\u002Ffixtures\u002Frules\u002FAG-SYS-002\u002Fpositive\u002Ffault\u002Fcsv-tool\u002FSKILL.md",[],[],{"skillNo":69,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":18},"01M7DJ7F00PA7EHFM39RVCC8R5","https:\u002F\u002Fgithub.com\u002Fak2dev\u002Fagentguard\u002Fblob\u002Fmain\u002Ffixtures\u002Frules\u002FAG-SYS-002\u002Fnegative\u002Fnormal\u002Fcsv-tool\u002FSKILL.md",[],[],{"skillNo":74,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":18},"01M7DJ7DP7KDW7MYY5Q0R0JFEN","https:\u002F\u002Fgithub.com\u002Fak2dev\u002Fagentguard\u002Fblob\u002Fmain\u002Ffixtures\u002Frules\u002FAG-SYS-001\u002Fpositive\u002Ftoo-large\u002Fcsv-tool\u002FSKILL.md",[],[],{"skillNo":79,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":18},"01M7DJ7C3VA0KQ8Q9ESNX6AJHE","https:\u002F\u002Fgithub.com\u002Fak2dev\u002Fagentguard\u002Fblob\u002Fmain\u002Ffixtures\u002Frules\u002FAG-SYS-001\u002Fnegative\u002Fsmall\u002Fcsv-tool\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":18},"01M7DJ7AQZTGY5YADQX73TRB7K","pdf","https:\u002F\u002Fgithub.com\u002Fak2dev\u002Fagentguard\u002Fblob\u002Fmain\u002Ffixtures\u002Frules\u002FAG-SKL-TYP-002\u002Fpositive\u002Fother-publisher\u002Fpdf\u002FSKILL.md",[],[],{"skillNo":90,"name":85,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":18},"01M7DJ7831XDVWS4W9SAAKBJHX","https:\u002F\u002Fgithub.com\u002Fak2dev\u002Fagentguard\u002Fblob\u002Fmain\u002Ffixtures\u002Frules\u002FAG-SKL-TYP-002\u002Fnegative\u002Fsame-publisher\u002Fpdf\u002FSKILL.md",[],[]]