A method for converting text to audio to improve a design of electronic mail messages includes converting, by a text-to-speech converter component executing on a first computing device, an electronic mail message into an audio file. The method includes analyzing, by an analysis engine executing on the first computing device, the electronic mail message, the audio file, and a transcript of the audio file to identify at least one attribute. The method includes receiving, by a recommendation engine executing on the first computing device, an identification of the at least one attribute. The method includes generating, by the recommendation engine, at least one recommendation for improving a design of the electronic mail message, based on the identification of the at least one attribute. The method includes providing, by the first computing device, to a second computing device, the audio file and the generated at least one recommendation.