1. Authentication (Important): You must pass your raw NestUI API key directly in the workflow action's properties. Do NOT add the "Bearer " prefix; just use the raw key itself.
2. Generation (Instant Action): Use the "Generate Audio (POST)" action in your workflow. Unlike image or video generation, this model is synchronous and will instantly return the output_audio_url. No status polling is required!

3. Required Parameters:

Prompt (Text): The text you want converted to speech.

Language (Text): Must be exactly one of the following: Chinese, English, German, Italian, Portuguese, Spanish, Japanese, Korean, French, Russian.

Voice (Text): Must be exactly one of the following: Cherry, Serena, Ethan, Chelsie, Momo, Vivian, Moon, Maia, Kai, Nofish, Bella, Jennifer, Ryan, Katerina, Aiden, Eldric Sage, Mia, Mochi, Bellona, Vincent, Bunny, Neil, Elias, Arthur, Nini, Seren, Pip, Stella, Eric, Peter, Roy, Marcus, Li, Andre, Emilien, Dolce, Alek, Sonrisa, Bodega, Jada, Rocky.

4. Storage (Crucial): Generated audio URLs are temporal and expire after 24 hours. You must save the generated output to your database (using Bubble's ":saved to S3" operator on the returned URL) to ensure permanent access.