A system and method for providing user input to an application executing on an interactive television system wherein a user provides non-textual information to the interactive television system and this information is converted by a server to textual information which is provided to the application. In one embodiment, a digitizer pad coupled to a set-top box is used to digitize the user's handwriting. The digitized information is conveyed to a remote server which converts the digitized handwriting data into textual information. The textual information is conveyed to the set-top box, where it is input to an application executing on the set-top box. In another embodiment, a microphone is coupled to a set-top box. The microphone allows the user to input voice information which is digitized and conveyed to the server for conversion into textual information. The textual information is conveyed back to the set-top box and is input to an application executing on the set-top box.