In state security and forensic contexts, auditory surveillance through wiretapping and sound recording is as old as the technologies that have enabled it since the 1890s. Historical and critical studies of systematic eavesdropping have commonly focused on the act of taping, the decoding of encrypted messages, and the sociopolitical contexts of those activities. This project, however, centers on the history of scientific and humanities-based research into the recordings’ sonic features—the characteristics of voices, speech, and non-speech sounds—for speaker identification. It studies two settings: the Ministerium für Staatssicherheit (Stasi) in the German Democratic Republic and its research program on sound; and the field of audio forensics in the United States. Which characteristics of sound did the German and American experts consider relevant? What analytic techniques did they use? How did forensics and state security programs inform each other? And how did these practices affect the use of sonic skills in the sciences and humanities more widely? Methodologically, the project draws on the analysis of archival documents and sonic elicitation interviews. Theoretically, it aims to link STS theories on “sonic skills” with work on “acousmatic sound.” And societally, it aims to put recent debates about auditory surveillance into historical context. One of the events in the Hidden Ears project is the closed expert workshop “Slicing Sound: Speaker Identification and Music Censorship in the Former Eastern Bloc” (October 2019), which will situate the history of Stasi speaker identification more deeply in GDR historiography as well as in sonic state security activities in other communist countries in the Cold War era.