Abstract: A passive personal vocal monitor is described for enhancing the sound of a singer's voice to the singer during a performance. The monitor has an earpiece with a nipple configured to be disposed snuggly within the singer's ear and having an internal channel terminating in an open end adjacent the singer's eardrum. A tubular acoustic waveguide extends from the earpiece to an open end that is positionable adjacent the singer's mouth when the nipple is in the singer's ear. A canal in the earpiece couples the tubular waveguide to the channel of the nipple so that sounds of the singer's voice are conveyed along the waveguide and to the user's eardrum during a singing performance. An adjustable ambient sound inlet is provided on the earpiece so that a singer may admit a desired amount of ambient sound to be mixed with his voice as he sings.