When we use our voice to communicate, it carries all sorts of meanings unrelated to the content of our message. Even in the absence of a face-to-face encounter, our interlocutors excel at extracting the information about our age, sex, gender and sexuality, regional background, native language(s), ethnicity, socioeconomic status and levels of education. Listeners can extract this information alone from the properties of our voice. While speech perception is a crucial part of any language use, our tendency to imbue speech with a wide range of social meanings opens doorway to biases, possibly leading to social inequality and injustice. In this seminar, we review existing evidence on social aspects of (primarily English) speech and the associated biases, with a special focus on latest experimental studies involving high-fidelity human-like avatar technologies.\u00a0We cover the bases for individual differences in human speech and voice and discuss which aspects are or are not under voluntary control, concluding with food for thought on the complex relationships that exist between language and society.\u00a0The seminar will be accessible to a broad interdisciplinary audience with an interest in sociolinguistics.<\/p>\n