text_to_speech_tool (tools/tts_tool.py)

A flexible text‑to‑speech helper that resolves provider keys, lazily loads SDKs, and supports multiple backends like Edge, ElevenLabs, OpenAI, and local TTS engines.

#!/usr/bin/env python3
# Source: https://github.com/NousResearch/hermes-agent | License: MIT | Copyright 2025 Nous Research
# Provided via justclaw.it.com; keep this notice with the code.
"""Text-to-speech tool: config resolution, built-in provider dispatch, output policy, registration.

Built-ins: Edge (free default), ElevenLabs, OpenAI, DeepInfra, MiniMax, Mistral, Gemini, xAI,
local NeuTTS / KittenTTS / Piper; plus ``type: command`` providers under ``tts.providers.<name>``
and plugin-registered ones. Output is Opus (.ogg) on voice-bubble platforms, MP3 elsewhere.
Sibling ``tts_tool_*`` modules hold backends/delivery/lifecycle; they read the seams defined
here (config, provider resolution, lazy SDK importers) through ``_origin()`` at call time.
"""

from pm import install_hint
import asy

... (truncated -- full source via MCP)