This paper aims to address a preliminary draft addressing a complex issue of social marker translatability in a case of Polish-English speech translation performed by a native speaker of Polish. However, this is not a case study: the aim is to bring together some findings and ideas concerning the nature of social marking in L1 and L2 production, accent perception and evaluation, phonetic interference and limits on phonetic accuracy in a second/foreign language speech production which are believed to be relevant to the problem of accuracy in speech translation.