Master 2013 2014
Stages de la spécialité SAR
Development of an acoustically driven system for traffic noise monitoring

Site :Fraunhofer IDMT, Hearing Speech and Audio Technology
Lieu :Oldenburg, Germany
Encadrant : Danilo Hollosi, Jens Schröder,
Dates :To be determined.
Rémunération :500€ Monthly
Mots-clés : Parcours ATIAM : Traitement du signal


In the European Union, cities with more than 100,000 inhabitants are obliged to measure the noise pollution and to publish them as noise maps. If these measurements indicate a violation of defined thresholds, actions for decreasing the noise pollution are supposed to be induced. Unfortunately, the common and standardized measurements and procedures are not able to distinguish between different kinds of noise sources. Thus, appropriate and efficient actions for noise-polution reduction cannot be applied. The noise maps have to be regarded as false. At the Fraunhofer Institute For Digital Media Technology, Project Group Hearing, Speech and Audio Technology, machine learning based approaches for acoustic event detection in urban environments are investigated. Therefore, an acoustic utterance recorded by microphones is transcribed into features representing the acoustic class or speech command. These features are fed to a machine learning method that trains models to automatically classify acoustic events. The goal of the proposed thesis is to develop a method to measure and quantify traffic noise pollution based on acoustic data and machine learning algorithms. The method is supposed to run on an embedded system. Thus, it has to be computationally efficient. For the evaluation, data of traffic volume and results from deterministic algorithms for traffic noise pollution will be provided. The main work lies in developing and implementing an adequate machine learning approach, the collection of acoustic recordings and their annotation, the development of a procedure for traffic noise pollution monitoring and the evaluation of the system.