PRODUCTION SCENARIO
A Foundry app generates written answers and must play them back to the caller as spoken audio.
Which Azure Speech capability does the app call?
Answering here is anonymous. Nothing is saved unless you sign in.
Show answer and explanation
Answer: Text to speech synthesis with a standard neural voice
Text to speech, also called speech synthesis, converts text into human-like spoken audio using a standard neural voice. Speech to text runs the other direction, turning the caller's audio into a transcript. Speech translation also starts from spoken input, which the app does not have. Batch transcription is an asynchronous job over audio files already in storage, so it neither starts from text nor returns audio.