The assessment and evaluation of speech quality and speech-output systems is a multi-dimensional problem. In this essay a broad overview of relevant methods and methodologies is given, and their range of validity is analysed. It shows that - although progress in the field can be noted - there is still a significant lack of knowledge and experience in many ways. Until now a number of important factors of speech quality have been identified, but reliable methods for monitoring the performance of speech quality and speech-output systems in praxi are hardly available. In other words, an integrated, application-oriented battery of assessment and evaluation procedures is still missing - as well as scientifically-based written guidelines to support non-experts when exposed to respective assessment and evaluation tasks. It is the aim of this contribution to review methods currently available in the field - including some recently novel approaches - in view of the task of predicting performance of speech-output devices in particular application scenarios.