clawmate-companion-tts

Use voice-first replies to automatically generate brief spoken responses when conversation tone signals that speaking is more effective than text. This improves emotional nuance and user engagement during intimate check-ins.

Use TTS instead of text for moments like bedtime chats, storytelling, comforting messages, and soothing chats where a spoken tone conveys nuance better than text. The workflow decides when to switch based on conversation signals.

Short voice messages must adhere to safety, length, and naturalness constraints. The deterministic workflow enforces these limits while generating concise spoken content to ensure appropriate and natural audio output.

Yes, the TTS skill operates without external dependencies, integrating seamlessly with the existing pipeline via the clawmate_generate_tts function. It prepares content internally before invoking the voice generation pipeline.

Start delivering voice messages by asking the personal assistant to generate a short voice message for the current reply. The skill evaluates the context and automatically invokes the TTS function with the prepared spoken content.