European Union

 

Funded by the Horizon 2020 Framework Programme of the European Union


15 Early Stage Researchers (ESRs) will be employed to undertake research in the framework of the project “VRACE: Virtual Reality Audio for Cyber Environments”, and will be funded for 36 months through the prestigious Marie Skłodowska-Curie Actions (MSCA) Innovative Training Network (ITN) programme. VRACE will establish a multidisciplinary training and research programme focusing on the analysis, modelling and rendering of dynamic 3-dimensional soundscapes for applications in Virtual Reality (VR) and Augmented Reality (AR), delivered by nine cooperating European Universities and their industrial partners, including Siemens, Mueller BBM, Sennheiser and Facebook Reality Labs (former Oculus). In addition to their individual scientific projects, all ESRs will benefit from further continuing education through a dedicated training program in the various fields of expertise of the consortium partners, which includes active participation in workshops, conferences and outreach activities.

Eligibility Criteria:

  • At the time of recruitment by the host organisation, be in the first four years (full-time equivalent) of their research careers and not yet have been awarded a doctorate. This four-year period is measured from the date of obtaining the degree that would formally entitle to embark on a doctorate.
  • Must not have resided or carried out their main activity in the country of the host organisation for more than 12 months in the 3 years immediately prior to their selection for this post.

ESR1 (Austria-MDW): Informed source separation based on 3D sound fields and laser-based techniques for remote vibration sensing. (WP1; link to application)

Objectives: Reconstruction of near field sound of multiple sources from wave field analysis and additional cues obtained from sensors and prior knowledge (e.g. 3D environment model). Study of laser-based techniques for sound source localisation and remote vibration sensing.


ESR2 (Austria-MDW): Spatial modelling of sound radiation of a range of musical instruments including wind instruments and grand piano. (WP1; link to application)

Objectives: Investigate methods to quantify experimentally the directional radiativity properties of a range of musical instruments, paying particular attention to the relevance of such measurements for real-world situations and applications. Use of microphone-array and motion capture techniques to collect sound radiation data for different playing styles and dynamic ranges. Use of these experimental data to validate and inform related physics-based and fully-embedded acoustic simulations in the time domain. Extract directivity functions from experimental data and numerical simulations which allow deriving a realistic near field sound based on dry music recorded in an anechoic chamber.


ESR3 (Austria-MDW): Collision modelling for sound synthesis of musical instruments and other vibrating objects. (WP1; link to application)

Objectives: Studying the nonlinear dynamics of collisions between elastic solids with the aim of modelling sound excitation mechanisms in musical instruments (such as the vibrating lips of brass players, the reeds of wind instruments or hammer-string / mallet-membrane interactions). The developed non-smooth numerical methods will be extended to a variety of sound-generating objects that may be encountered in VR / AR.


ESR4 (Austria-TUW): Parallelization of scalable domain decomposition methods for GPU architectures (WP3, WP2; link to application)

Objectives: Studying and improving convergence of iterative hierarchical solvers. Optimization of scalability and possible degree of parallelism. Proof of concept and benchmarking of investigated methods in terms of memory requirements, computing performance and scalability on GPUs


ESR5 (Germany-HfM): Complex sources and musical ensembles – psychoacoustic assessment of virtual sound fields. (WP4, WP2; link to application)

Objectives: Investigation of the interaction of small ensembles such as quartets with variable room acoustic conditions. Investigation of the effect of variable reverberance and reflections on the performance of ensembles. Studying how individual and joint performance strategies of musicians are affected by different acoustic environments. Wave field analysis and synthesis of orchestra and ensemble sounds with sensory evaluation of quality parameters of analysis and reproduction.


ESR6 (Netherlands-TUe): Audio-visual timing tolerances - sensory evaluations of different stimulus properties (WP4; link to application)

Objectives: Experimental determination of the JND in the perception of asynchrony between auditory and visual components of a media event and its modalities. Determination of the relationship between perceived asynchrony and VR immersiveness.


ESR7 (France-CNRS): Physics-based outdoor sound synthesis for moving sources in realistic environments (WP2, WP1; link to application)

Objectives: Adapting existing physics-based emission models for moving sources (vehicles or wind turbine blades) and for background noise (wind-induced noise and vegetation noise) in a sound synthesis context. Accounting for the effects of atmospheric variations, ground and obstacles in the propagation model of the sound synthesis tool.


ESR8 (Belgium-KUL): Hybrid methods in room acoustics - efficient combination of numerical models for full audio range simulations for virtual reality (WP2; link to application)

Objectives: Development of a single sound propagation model that is valid in the full audible range, based on a combination of element-based approaches and geometrical methods, using parallel filter banks for different frequency bands, taking GPU acceleration in consideration.


ESR9 (Belgium-KUL): HRTF modelling - efficient, binaural rendering for individual virtual acoustic environments (WP3; link to application)

Objectives: Efficient numerical calculation of HRTFs for individual VR experiences based on mesh generation techniques for head/torso/pinna such as photogrammetry and model-order reduction techniques. Investigating suitability of different methods for hardware acceleration on GPUs.


ESR10 (Finnland-AALTO): Room discrimination thresholds - determination of listeners' tolerance to imperfections of virtual acoustic environment (WP4; link to application)

Objectives: Experimental determination of the JND (just noticeable differences) of certain parameters of spatial soundscapes and thresholds beyond which degradation of sound source localisation, disorientation and possible motion sickness starts to occur.


ESR11 (Germany-LUH): Binaural rendering - optimizing acoustics and signal processing methods for binaural synthesis in virtual and augmented reality (WP3; link to application)

Objectives: Efficient, adaptive, realistic and good sounding binaural synthesis for VR/AR, Acquisition, comprehension and compensation of headphone influences. The IKT will contribute with its competence in the field of signal processing for virtual acoustics and the assessment of subjective aspects of human hearing in listening experiments. There will be a strong collaboration with the ESR at SENN especially in the area of headphone related acoustics.


ESR12 (UK-QUB): Physics-based source modelling with time-variant parameters for immersive audio (WP1; link to application)

Objectives: Introducing extended parameter time-variance in physics-based simulations of musical instruments and other mechano-acoustic sources. Studying stability and energy conserving properties of numerical algorithms with respect to dynamic parameter variations. Sound synthesis under time-variation of parameters as articulated by musicians and other users.


ESR13 (Germany-BBM): Advanced room acoustics modelling methods and validation of predicted or measured quality parameters (WP2; link to application)

Objectives: To study, compare and further develop advanced acoustic modelling and auralisation methods for interior spaces. Assessment of the subjective relevance of predicted or measured quality parameters for concert theatres, multipurpose halls and other communication spaces. Use of sensory focussed approaches to identify the auditory relevance of architectural features.


ESR14 (Belgium-SISW): GPU accelerated Discontinuous Galerkin Methods for real-time acoustic simulations. (WP2; link to application)

Objectives: Studying Discontinuous Galerkin Methods to find the most efficient ways of implementing them on massively parallel hardware. Studying CUDA, OpenCL, and OpenAAC as a base for personal high performance parallel computing. Developing applications in structural acoustics, room acoustics and environmental acoustics. Comparing computing times for different software and hardware platforms and architectures.


ESR15 (Germany-SENN): VR Headphone technologies - influence of headphone characteristics on realism and sound localisation in VR (WP3; link to application)

Objectives: Numerical simulations of the sound generation in the speaker and the sound transmission from speaker to ear drum. Optimization of the acoustical transparency of the headphone by means of active feed through. Active compensation of occlusion effects. Adaptive compensation of wearer or device related parameter variations. Incorporating human perception models for the optimization of headphone characteristics. Validation of the results by sensory evaluations.