The proper operation of hydroelectric generating units, one of the core pieces of equipment in a hydroelectric plant, is critical to the safe and stable operation of the plant. Abnormal sound detection in the power plant could prevent accidents, improve efficiency, save costs and ensure the safety of personnel. Thus, hydropower plants should attach great importance to abnormal sound detection, strengthen equipment maintenance management to ensure safe and stable production operation. At present, the sound detection system in hydropower plants is mainly used to monitor the operating status of equipment such as hydraulic turbines, generators, bearings, gears and pumps. These devices produce sounds of various frequencies and amplitudes during operation, and the system can analyse the characteristics of these sounds to determine whether the equipment is operating normally. Abnormal sounds are usually caused by equipment failure or damage, such as turbine blade damage, bearing wear, poor gear mesh, etc. These abnormal sounds are usually different from the sound characteristics of the equipment during normal operation and can be identified and monitored by the sound detection system. Abnormal sound detection technology mainly includes the following steps, i.e., sound signal acquisition, signal preprocessing, feature extraction and fault diagnosis. Among them, feature extraction is the core of abnormal sound detection. By extracting and analyzing the spectrum, waveform, energy and other features of the sound signal, it can determine whether there is abnormal sound in the equipment. This technology has been widely used in the production practice of hydroelectric power plants and has the advantages of high reliability, high detection accuracy and high realtime performance. The main purpose of this paper is to give a brief overview of abnormal sound detection technology for hydroelectric generating units, discuss and summary the latest researches.