Audio version — Estimated duration: 1 min 33 sec
Convert markdown documentation into spoken audio with independent LLM and TTS backends — mix and match online and local engines.
Features
- Dual backends — Cerebras API or llama.cpp for LLM; edge-tts or Piper for TTS
- Batch mode — Recursively process an entire
.mddirectory tree - Single mode — Convert one file with a single command
- Audio embedding — Injects
<audio>tags back into source files with duration metadata - Conversational output — LLM rewrites docs into spoken-word text before synthesis
Backend matrix
| Component | Online | Local |
|---|---|---|
| LLM | Cerebras API | llama.cpp server (localhost:8080) |
| TTS | edge-tts (Microsoft cloud) | Piper (GPU via onnxruntime) |
Quick nav
| Section | What’s there |
|---|---|
| Installation | pip install, API key, local server setup |
| Quickstart | Generate your first audio in 2 commands |
| Pipeline Mode | Batch-process a directory of .md files |
| Single File Mode | Convert one .md file |
| CLI Reference | Every argument explained |
| Backends | Online and local LLM/TTS setup guides |
| Output | File layout, formats, audio embedding |
| Architecture | Pipeline internals, rate limiting, ordering |
| Troubleshooting | Common errors and fixes |
Tech stack: Python 3.10+ · OpenAI SDK · edge-tts · Piper · onnxruntime-gpu · Cerebras API · llama.cpp