The service provides speech synthesis by using text input to generate voice output. Real-time conversion takes text input to the service and outputs the content as audio. Vocal inflections are designed to match human voice, to the output sounds like a person reading the screen text aloud.
API methods support input of text as part of the request and conversion to create mp3 or wav audio files, which then play through system speakers. Methods allow selection of language and accent, with choices including British English, French, Spanish as spoken in many parts of Spain, and both European and Brazilian Portuguese.