Question

You are building a voice-enabled customer support agent that must speak responses aloud in a distinctive, consistent brand persona that does not match any of the prebuilt Azure voices. Legal has approved using recordings from a professional voice talent who signed a consent form. Which Azure AI Speech capability should you use to generate the agent's spoken output?

Accepted Answer

Train and deploy a Custom Neural Voice (professional) model from the voice talent's recordings and use it as the text-to-speech voice

Answer

Configure text-to-speech with a prebuilt neural voice and adjust SSML prosody rate and pitch to approximate the brand persona

Answer

Use Custom Speech to train an acoustic model from the voice talent's recordings so responses match the brand persona

Answer

Use the Speech translation API to convert generated text into the target voice persona