Module tts 

Multi-provider Text-to-Speech (TTS) subsystem.

Supports OpenAI, ElevenLabs, Google Cloud TTS, Edge TTS (free, subprocess-based), and Piper TTS (local GPU-accelerated, OpenAI-compatible endpoint).

per-instance configs live under [tts_providers.<type>.<alias>]; agents pick which instance to use via the tts_provider dotted alias reference. Global runtime knobs (default_voice, max_text_length, etc.) live on [tts].

Structs

EdgeTtsProvider

Edge TTS model_provider — free, uses the edge-tts CLI subprocess.

ElevenLabsTtsProvider

ElevenLabs TTS model_provider (POST /v1/text-to-speech/{voice_id}).

GoogleTtsProvider

Google Cloud TTS model_provider (POST /v1/text:synthesize).

OpenAiTtsProvider

OpenAI TTS model_provider (POST /v1/audio/speech).

PiperTtsProvider

Piper TTS model_provider — local GPU-accelerated server with an OpenAI-compatible endpoint.

TtsManager

Central manager for per-agent TTS synthesis.

Traits

TtsProvider

Trait for pluggable TTS backends.