SPEC.md — Audio Generation Scripts Scope Thin orchestrator scripts delegating TTS synthesis to codomyrmex.audio.text_to_speech.Synthesizer. Inputs Input Source Type Provider config/audio/config.yaml or --provider "edge-tts" or "pyttsx3" Voice config/audio/config.yaml string (Edge TTS voice ID) Rate / Pitch / Volume config/audio/config.yaml float Text config/audio/config.yaml or --text CLI arg string Batch texts config/audio/config.yaml list[string] Output dir config/audio/config.yaml path string Outputs Output Location Format Single synthesized audio output/audio/orchestrate_single.mp3 MP3 (edge-tts) or WAV (pyttsx3) Batch audio files output/audio/orchestrate_batch_N.mp3 MP3 or WAV Console logs stdout CLI helper format Exit code process 0=success/skip, 1=error Error Handling Condition Behaviour Audio extras not installed print_warning + exit 0 Primary provider unavailable Try fallback provider Both providers fail print_error + exit 1 Synthesis fails print_error + exit 1