What type of inputs will we need for expressive speech synthesis?

Speech synthesis is not necessarily synonymous with text-to-speech. This paper describes an implementation for a talking machine that produces multilingual conversational utterances from a combination of speaker, language, speaking-style, and content information, using icon-based input. The paper addresses the problems of specifying the text-content of a conversational utterance from a combination of conceptual icons, in conjunction with language and speaker information. It concludes that in order to specify the speech content (text details and speaking-style) adequately, further selection options for speaker-commitment will be required.