Abstract: For improved sound reinforcement, especially of the human voice, in situations where clarity and full audience coverage are important, an integrated structure supports a microphone and a speaker unit in an anti-feedback spatial relationship. The single speaker unit, mounted in an elevated position facing the audience, acts as a well-localized "point source", the microphone being located nearby in a working location to the rear of speaker. Despite this proximity, acoustic feedback and resultant "howl" tendencies are suppressed by providing both the microphone and the speaker unit with hypercardioid directivity and locating them in an offset back-to-back relationship such that alignment of their "null" regions combines their attenuation, minimizing feedback. In one embodiment, a lectern table supports the microphone in a working position and mounts an offset vertical pole carrying the speaker unit. The system amplifier may be housed within the lectern table.