Problem or Use Case

Add official support for OpenRouter's new text-to-speech API which features models such as GPT-4o Mini TTS, Voxtral Mini TTS and Gemini 3.1 Flash TTS Preview.

Motivation

• While some of the models available are already supported via other providers, Users who already have their OPENROUTER_API_KEY set up can enable TTS with no extra accounts, keys or setting up billing
• Compatible with OpenAI TTS API which is already supported so it can be possibly implemented without a lot of code changes

Docs

Official guide: https://openrouter.ai/docs/guides/overview/multimodal/tts
API reference: https://openrouter.ai/docs/api/api-reference/tts/create-audio-speech

Proposed Solution

• Use OPENROUTER_API_KEY and existing openrouter authentication flow, either one for LLMs or tools/openrouter_client.py
• .hermes/config.yaml schema:

tts:
  provider: "openrouter"
  openrouter:
    model: "openai/gpt-4o-mini-tts-2025-12-15"
    voice: "alloy"
    speed: 1.0 # only supported on some models

Alternatives Considered

No response

Feature Type

Other

Scope

None

Contribution

• I'd like to implement this myself and submit a PR

Debug Report (optional)