A method and system for Extensible Markup Language (XML) application transformation may include converting a call flow diagram describing a voice interface process into a list of states in a XML format, and creating a lookup table of audio states in the XML format by mapping a plurality of audio prompts and their corresponding textual representations with states of a list of states that play audio files associated with the plurality of audio prompts. The method and system may include creating an intermediate application in the XML format and from the list of states by merging audio prompts in the lookup table with states of the list of states that play audio files, and transforming the intermediate application into a second application of a second format that is a representation of the call flow diagram.