skills.rest
lettucebo
lettuceboCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

podcast-generation

Generate podcast-style audio narratives from text using Azure OpenAI Realtime API over WebSocket.

Updated Mar 25, 2026
Search Tags:#react#fastapi#text-to-speech#podcast#audio generation#websocket#azure openai
One-click install
npx skills add https://github.com/lettucebo/Skills --skill podcast-generation-lettucebo
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: podcast-generation
Source: https://github.com/lettucebo/Skills/tree/main/skills/microsoft/podcast-generation
Command: npx skills add https://github.com/lettucebo/Skills --skill podcast-generation-lettucebo

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires openai, and includes scripts (resource) and references (resource) components.

What problem does it solve? Converting written content into spoken audio requires wiring together a real-time speech API, streaming audio chunks, and format conversion, which is complex to implement from scratch. This Skill provides a complete full-stack pattern for generating podcast-style audio narratives with Azure OpenAI's GPT Realtime Mini model. ## Core Features & Use Cases - Realtime Audio Generation: Connects to Azure OpenAI Realtime API via WebSocket, streams PCM audio chunks and transcripts, and converts them to WAV for playback. - Full-Stack Implementation: Covers a Python FastAPI backend with database persistence and a React frontend that plays base64-encoded WAV audio. - Voice and Style Control: Supports six voices (alloy, echo, fable, onyx, nova, shimmer) and narration styles such as podcast, summary, and lecture. - Use Case: A bookmarking app lets users select a tag or article and click "Generate Podcast" to hear a 1-2 minute spoken narrative summarizing the saved content. ## Quick Start Ask the AI to generate a podcast-style audio narrative from a piece of text content using the Azure OpenAI Realtime API with the gpt-realtime-mini deployment.

Frequently Asked Questions about podcast-generation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate audio with the Azure OpenAI Realtime API?▼

Connect via WebSocket using the OpenAI Python client's realtime.connect method with the gpt-realtime-mini model, set output_modalities to audio, send a text prompt, and collect response.output_audio.delta events containing base64 PCM chunks.

How to convert PCM audio to WAV in Python?▼

Use the pcm_to_wav script, which writes a RIFF/WAVE header around raw PCM bytes with the correct sample rate, channels, and bit depth. Azure Realtime API outputs 24kHz, 16-bit, mono PCM, so pass sample_rate=24000.

What voices does Azure OpenAI Realtime API support?▼

The skill documents six voices: alloy (neutral), echo (warm), fable (expressive), onyx (deep), nova (friendly), and shimmer (clear). Select one via the voice_name parameter when generating the narrative.

Why is my Azure OpenAI Realtime WebSocket connection failing?▼

The endpoint must be the base URL without /openai/v1/ in the environment variable; the code converts https:// to wss:// and appends /openai/v1 itself. Also verify AZURE_OPENAI_AUDIO_API_KEY and the deployment name are set correctly.

How do I play base64 WAV audio in a React frontend?▼

Decode the base64 string into a Uint8Array, create a Blob with MIME type audio/wav, generate an object URL with URL.createObjectURL, and assign it to an Audio element to play it in the browser.

Related Skills

Explore More Software Engineering Skills→
istota-projectistota-project
10

bookmarks

Search, browse, and manage Karakeep bookmarks via the Istota API.

Official
Advanced
microsoftmicrosoft
2.9k

podcast-generation

Generate podcast-style audio narratives from written content via WebSocket streaming.

Official
Advanced
ax-llmax-llm
2.9k

ax-cpp-audio

Map C++ audio input and realtime events through the axllm provider API.

Official
Intermediate
Umbral-BotUmbral-Bot

azure-audio

Generate text-to-speech audio via the Azure OpenAI Realtime API.

Official
Intermediate