Manually trigger text-to-speech with customizable voice and speed
Call the MCP tool `text_to_speech` with the following parameters:
- text: {text}
- voice: {voice}
- speed: {speed}
- format: "base64_wav"
The MCP tool will synthesize speech and automatically play the audio.
If the text is longer than 800 characters, warn the user and truncate it.
Available voices:
- American Female: af_bella (warm), af_sarah (professional)
- American Male: am_adam (calm), am_michael (confident)
- British Female: bf_emma (refined), bf_isabella (elegant)
- British Male: bm_george (distinguished), bm_lewis (articulate)
After calling the tool, briefly confirm what was spoken.