Moderate donation TTS with AI

Return a speech decision, moderated text and a selected language.

Before you start

Use a test project connected to Core. Configure the relevant integrations in the intended local service profile. Review Apply and Save before testing a changed definition.

Steps

1. Configure the donation signal, an AI model profile and Piper voices.
2. Create a reusable moderation macro with donation text as an explicit input.
3. Ask Call and Store for a structured decision with moderated text and a supported language/voice choice.
4. Parse and validate the result before selecting Piper Speak; handle malformed or missing fields as a deliberate no-speech/error path.
5. Call the macro from the donation reaction and map its output explicitly.
6. Save the automation and Test allowed text, disallowed text, empty input, failed model calls and unsupported language.

Verify

Only a validated permitted result reaches the chosen speech action. Inspect the corresponding run, then Save the applied project when the result is correct.

If it does not work

AI output is not a guarantee of moderation accuracy. Keep a stop/mute control, avoid unlimited retries and preserve a fallback for unavailable provider/voice choices.