Hermes LoopHermes Loop
Media · text-to-speech

Text to speechElevenLabs primary. OpenAI-compatible /audio/speech is fallback.

← Media
What this is
Sandbox surface for the text_to_speech tool.
What to do
Type text, optionally pick a voice, hit Synthesize. Audio is returned base64-encoded and previewed inline.
What happens next
When agents call this tool inside missions, each call needs an approval before it runs.
ElevenLabsActive voice ID: 21m00Tcm4TlvDq8ikWAM(default — set ELEVENLABS_VOICE_ID to override)
Optional. Leave blank to use the active voice ID. Override per-call by pasting any ElevenLabs voice ID.