The purpose of this study is to learn how to improve synthesised speech. In order to do this we are gathering people's opinions on the output of several different speech synthesis systems. You will be performing a series of evaluations of the speech produced by these systems. In each one you will be asked to play audio files in order to listen to some sentences and you will be asked to give your opinion about some aspect of what you hear. It is important to understand that it is NOT you who is being tested - your results only provide information regarding the systems we are evaluating!

You must complete the whole evaluation, but you don't have to do it all in one go. If you leave it and then login again, you will be taken to the point where you left off. It should take around 1 hour to complete all the sections. No section or part of a section can be repeated - once you have submitted a response it cannot be changed. Please listen to the audio files using headphones or good quality speakers. Note that what we want you to concentrate on in each section may be different, and the way of entering your response will vary. In some sections you will select a description or score from a list or on a horizontal slider, and in others you will type in what you hear. Detailed instructions for individual sections will be repeated at the start of each section.

After you have completed the evaluation, you will be directed to a feedback questionnaire. This will give you an opportunity to tell us what you thought of this study. We would also like to gather some relevant demographic information, but please note that all questions in the feedback questionnaire are optional.

This is a research study. Your participation is fully voluntary and anonymous. Your email address will not be used except to keep track of your progress, so you can return to the same position later, should you require more than one session to complete all the sections.