The aim of the MADRAS project is to develop a new generation of instruments which can automatically identify and quantify, in real time, the various acoustic sources which make up a given acoustic environment that people are exposed to? such instruments will enable noise annoyance in daily life and in the workplace to be better controlled.
Artificial intelligence techniques will be developed which will allow the automatic identification and quantification of the various acoustic events - such as the passing of a lorry or the sounding of a car horn - which occur during the period of analysis. The automatic source recognition algorithms used will operate by comparing the processed sound pressure level data in the time and frequency domains with parameters derived from a wide-ranging and comprehensive data base of typical noises.
The final instruments, which will be use for either on-site measurements or long-term monitoring, will be able to correctly identify the vast majority of sources which make up typical acoustic environments in daily life or in the workplace. The nature of each source will be identified with as much detail as possible - so as to differentiate, for example, between high speed passenger trains or heavy good trains. The instrument will, thus, potentially be able to identify in total, many hundreds of different kinds of acoustic events. Source identification will be possible even when the sources are corrupted by other noises.
The instrument will also provide the user with objective information about the acoustic sources identified in order that their acoustic impact on people can be quantified. Information will be given, for example, about the frequency, duration, and weighted sound pressure levels of the various events identified. The result will be a complete and detailed picture of the acoustical environment being investigated.