Back to the catalogue
SpeechChat
Text to speech
Reads an answer back aloud, in chat or as a speech step inside a flow.
Every assistant message carries a play button, and the audio is synthesised on demand rather than for every answer — so a conversation nobody listens to costs nothing extra.
Inside a flow, a text-to-speech step produces an audio file as a run artefact: useful when the output is an announcement, a phone prompt, or a summary somebody will listen to on the way to work.