Respiratory sounds (RSs) recorded from the chest and trachea are nowadays being electronically analysed by many investigators with a view to (i) determining the mechanisms of their production, and (ii) to develop automated diagnostic systems based on RS analysis, that objectively categorise RS as being associated with health or respiratory diseases. However, one problem that hampers this type of research is that almost every RS investigation team uses different equipment, protocols and analysis methods which, to varying degrees, makes inter-investigator results difficult to compare. The review first discusses the many variables involved in RS recording and analysis, and the different approaches used by different investigators, to highlight this problem and its consequences. Secondly, although the review cannot propose immediately acceptable guidelines and standards for RS analysis, it proposes a 'seed' set of guidelines that are 'up for discussion' between investigators in the field, the final goal being to inject a degree of standardisation in equipment and methods that are acceptable to all involved.