DecoWind is a 3-year Danish research project whose goal is to devise advanced control strategies for wind turbines and farms for minimizing their acoustic impact. Noise propagation models (Nord2000 and WindStar) are verified through dedicated measurement campaigns onshore. Long-range offshore measurements are also conducted to understand these specific conditions. A Parabolic Equation method for noise propagation, which have mostly been restricted to academic use in the past, is integrated into an engineering context for wind farm control. This framework can be used to define a wind farm optimal control strategy. The energy production is maximized while limiting the noise impact at dwellings, depending on the considered site, by operating the turbines using their different noise operational modes. This framework is demonstrated through numerical test-cases. In addition, a public survey is conducted to assess the socio-acoustical impact of wind turbine noise, looking at several factors. The ultimate goal is to produce a set of recommendations regarding wind turbine noise regulations that would connect the new engineering design capabilities and the findings regarding public annoyance. The project is a collaboration between DTU, Siemens-Gamesa Renewable Energy, FORCE Technology, and EMD International. In this contribution, the main achievements of the project are summarized.

The objective of the FAMOS project was to develop a European guidebook with practical applications for the National Road Administrations on how to handle noise annoyance by non-acoustic moderators in planning of roads. When all technically feasible and economically possible measures to reduce the noise have been used, there can still be need for further reduction of the annoyance. Analyses reveal that only 1/3 of the variance in the annoyance response is caused by the noise level. 2/3 are determined by so-called non-acoustic factors. The annoyance response therefore can be altered without changes to the actual noise level. The project searched moderators that have a large impact on the annoyance without changing the noise level. The search was an international literature survey on annoyance studies. Moderators found were: •Expectations to and visual appearance of noise barriers •Presence of vegetation/greenery •Access to a quite façade •Neighbourhood soundscape •Attitudes towards authorities and road owners •Perceived traffic safety Experimental tests using sound walks, questionnaires and listening tests have been performed to quantify some moderators. Various methods have been used to find, extract and analyse data and turn the results into models for moderators formulated for practical use.

The perceptual differences between the sound reproductions of headphones were investigated in a pair-wise comparison study. Two musical excerpts were reproduced over 21 headphones positioned on a mannequin and recorded. The recordings were then processed and reproduced over one set of headphones to listeners, who were asked to evaluate their perceived degree of dissimilarity. The two musical excerpts were used in separate experiments. The processing of the recordings consisted of compensating for the influences of the playback headphones worn by the listeners as well as for the mannequin's ear canals. A multidimensional scaling analysis revealed two dominating perceptual dimensions used by the listeners to differentiate the reproductions of the headphones. These dimensions were similar for the two musical excerpts. Objective metrics are proposed to describe them, leading to correlations ranging from 0.89 to 0.97 between the dimensions and metrics. The first perceptual dimension was associated with the relative strength of bass, while the second dimension was related to the relative strength of the lower midrange.

In this study the characteristics of compact loudspeakers in a stereo setup positioned in a standardized listening room were investigated. Perceptual evaluations of eleven loudspeakers were conducted on the basis of six selected sensory descriptors, chosen by experienced listeners during consensus meetings. Based on an analysis of the perceptual evaluation data, four of the descriptors were found suited for modeling, with the purpose of developing metrics for prediction of Bass depth, Punch, Brilliance, and Dark-Bright respectively. Bass depth and Punch were modeled as one due to high correlation between the two. The experimental setup included loudspeaker spinners, enabling fast positioning of loudspeakers. The prediction models were based on binaural recordings, processed using a loudness model, and developed on the basis of previous work on headphone modeling [1,2]. They were trained on a subset of the data (66%) and validated on the rest. The resulting metrics had high correlations with the perceptual ratings of the validation dataset (r = 0.85-0.96).

A method for rendering virtual sources is proposed, where the sources are perceived to be closer than the loudspeaker array radius. The development is based on an informal finding about closer perception of the range of virtual sources rendered coherently over multiple loudspeakers. To avoid sound quality issues inherent with coherent rendering, the input is split to two streams, one with more transients and the other with smoother temporal envelope. The transient stream is rendered over coherent reproduction, and the continuous stream is processed with timefrequency-domain spreading technique. The results from localization tests with moving sources show that the proposed method produces the perception of closer distances on both sweet-spot and off-sweet-spot listening.