HUT ACOUSTICS LAB
3-D SOUND GROUP
3-D Sound Group - Personnel and Projects
Virtual AcousticsA quick look into the world of virtual acoustics can be found in a form of a web presentation here. Unfortunately it is currently only in Finnish.
IntroductionAuralization is the process of rendering audio data by digital means to achieve a virtual three-dimensional sound space. Auralization and binaural processing of audio data is of great interest in several research fields: Virtual reality, Multimedia, Auditory displays, Sonification, Graphical user interfaces, Teleconferencing, Room acoustics simulation and Music technology. The 3-D Sound Group at the Acoustics Laboratory of the Helsinki University of Technology works in co-operation with the DIVA Group and explores different fields related to spatial hearing and auralization in real and virtual environments. Current projects concentrate on evaluation of different Head-related transfer function (HRTF) measurement techniques, real-time digital reverberation algorithm design, and DSP processing methods for headphone (binaural) and loudspeaker (transaural) reproduction.
The 3-D Sound Group has been involved in the design and implementation of the DIVA Virtual Audio System, which is a real-time virtual concert performing environment. We have implemented both realtime and non-realtime auralization methods on Texas Instruments TMS320C30 and TMS320C40 floating-point signal processors, and on Silicon Graphics workstations. The software of the signal processor based system has been created using Lisp-based QuickSig and QuickC30/C40 DSP environments that have been designed at the Acoustics Laboratory. The SGI workstation software for room acoustics and auralization has been developed by Lauri Savioja. The current signal procssor based auralization hardware consists of a 8x8-matrix AD/DA converter and a multiprocessor TMS320C40 system running on a Apple Macintosh Quadra 950. Future plans include the use of SGI 02 computers with the new 8-channel audio interface.
HRTF MeasurementsA Head-Related Transfer Function describes the free-field transfer function to the entrance of the human ear canal. Measurements of HRTFs can be carried out using miniature probe microphones fitted in the test persons ear canal (open ear canal measurement) or by using miniature condenser microphones fitted in the blocked ear canal. We have conducted several HRTF measurement series on test persons using Sennheiser KE211-4 miniature microphones. We have also performed measurements on dummy heads (Neumann Ku81i, B&K 4100, B&K 4128, Cortex MK2).
HRTF processing involves different digital filter design and approximation techniques. To achieve real-time performance on signal processors the HRTF data requires preprocessing and advanced filter design techniques. Current filter implementations are based on auditory resolution FIR or IIR filter design and smoothing techniques.
Auralization TechniquesThere are several approaches to the reproduction of auralized audio data. The two main categories are headphone and loudspeaker auralization. The main advantages of headphone auralization are the stability of the listener relative to the headphone placement and the low-cost testing environment (no need for an anechoic chamber). The main disadvantage is the instability of the HRTF measurement and the headphone equalization and the known problems in headpohone localization of sounds. The main disadvantage of loudspeaker auralization is the critically limited listening position and a need for a highly absorbing (nearly anechoic) listening environment.
The DIVA EnvironmentA real-time virtual audio reality model has been created. The system includes model-based sound synthesizers, geometric room acoustics modeling, binaural auralization for headphone or loudspeaker listening, and high-quality animation. The design goals of the overall project have been to create a virtual musical event that is as authentic as possible both in terms of audio and visual quality. Novelties of this system include a real-time image-source algorithm for rooms of arbitrary shape, shorter HRTF filter approximations for more efficient auralization, and a network-based distributed implementation of the audio processing soft- and hardware.
Related ProjectsMany research topics carried out at HUT Acoustics Lab relate to spatial hearing and 3-D sound. Examples of such research projects are: Virtual Sound Source Positioning Using Vector Base Amplitude Panning (Ville Pulkki), Sound Quality, Development of Models for Binaural Hearing, Research on Psychoacoustics (Prof. Matti Karjalainen), Binaural Recording and Reproduction (Panu Maijala), Listening Tests and Room Acoustics (Antti Järvinen).
Last Update: Feb. 25, 1997