TASCAR is an open-source software platform for real-time interactive virtual acoustic environment rendering for research and music applications. It is licensed under GPL-2.0. The source code is hosted at https://github.com/HoerTech-gGmbH/tascar/. Binary packages are provided by Hörzentrum Oldenburg gGmbH. For the current user manual (pdf) see manual.pdf. A video tutorial about setting up a session in TASCAR can be found here.
add frequency weighting and time constant to level2osc plugin.
add parenting feature to objects, fix bug in nested XML files, fix bug in nested TOCS files, fix segmentation fault on unload, improve pos2osc reliability, and other minor bugfixes and improvements.
improvements of windows build, systematic OSC documentation
preliminary windows build, scheduled OSC messages, improved scattering model, improvement of FDN based reverberation, band-level measurement plugin, many bugfixes and minor improvements
New plugins, many minor bug fixes, improvement of OSC handling, SVG map export via OSC, and minor improvements.
Correct FIR filter length in speaker correction (affects FIR speaker correction only, not related to calibration wizard).
Added plugin to generate ramps on changes of transport state, bugfix of FIR loudspeaker frequency correction, bugfix and improvement of LSL handling in the interface to Qualisys track manager, many minor bugfixes and improvements.
Bugfix and documentation of Matlab/Octave tools.
Bugfix of loudspeaker calibration wizard: When frequency-dependent speaker correction was active, the broadband calibration is now measured with active speaker correction.
Allow to use material names instead of filter coefficients in reflectors. Some materials definitions are built into TASCAR, more can be defined within a scene. Furthermore, the speaker calibration uses now the more robust Nelder-Mead method for optimization of parameteric equalizers.
This is a minor bugfix release to fix the behavior of the sndfile plugin: The timeline-independent playback of sound files did not work correctly when the file load was triggered via OSC. This, and the documentation of the sndfile plugin, is updated with this release.
The dataloggin supports a headless mode for command line applications, improved handling of LSL streams and some other improvements (e.g., support of datalogging on macOS). The GUI was partly re-designed, and works again on macOS and Windows. Head tracking with the ovheadtracker module was improved further.
The crossover filters in subwoofer bass management are now using correct Butterworth filter design with 24 dB/Octave slopes and corect phase. Furthermore, the calibration method was improved to avoid compensation of the crossover filters by the calibration process. Also the MPU6050 based head tracking was improved to achieve a resolution near 1/100 of a degree (compared to approx. 1 degree in the previous version).
The loudspeaker calibration wizard was completely redesignd, and spectral correction of loudspeakers based on parametric filter banks was added (optional). Also the data logging was improved and many other minor bugs fixed as well as features added.
With this release, some new features were added and some bugs were fixed: The spawning of subprocesses was redesigned for more control, e.g., re-spawning after process ended. The data logging was bugfixed; in some conditions the logging of LSL streams ended after several minutes. A method to discover OSC variables and their current values was added, which enables the development of GUIs. The offline rendering tools now support specification of channel maps (thanks to A. Moore), and the build on macOS was significantly improved.
Bugfix of diffuse sound field decoding: an invalid scaling of the decoding order gain resulted in lower spatial resolution. With this bugfix, the diffuse field processing of the directional mask plugin multibeam was implemented and tested.
Optional convolution of loudspeaker channels with HRIR for all speaker-based receivers.
Fix orientation and channel order of diffuse sound fields, minor bugfixes and documentation.
Fix dependency problem with LSL, minor improvements of various plugins, documentation and scripts.
Add watchdoc functionality to glabsensor jackstatus.
Fix of severe bug in tascar_spkcalib. If you are using 0.217 then please update immediately.
Microphone array receiver type, many minor bugfixes and improvements.
Support of subwoofer definition and calibration in speaker based render methods.
Bugfix of the hoa3d receiver type (tolerance value of pseudo-inverse), and many minor improvements and bugfixes.
Release of TASCAR for ARM Linux / Raspbian Buster.
A parametric HRTF simulation binaural receiver type was added. The parametrization is based on the Spherical Head Model (SHM) by Brown and Duda (1998) and includes three further low-order filters as originally suggested by Oliver Buttler and Stephan D. Ewert and implemented in the room acoustics simulator RAZR (Wendt et al., 2014; www.razrengine.com). This receiver implementation is an outcome of the bachelor thesis of Fenja Schwark.
Source directivity defaults now to omni-directional for volumetric (diffuse) receiver types. This can be overwritten for new source directivity types, e.g, to account for average diffuse radiation frequency response.
A partitioned convolution for for rendering diffuse reverberation based on recorded B-format impulse responses was added, and the simple feedback delay network (FDN) was improved.
A receiver type for rendering to ITU 5.1 speaker layouts was added. Furthermore, a new diffuse reverberation concept was developed, to allow diffuse reverb also in offline simulation and without recursive signal routing. A simple feedback delay network (FDN) was added as a prototype algorithm.
Addition of A weighting to level metering, some GUI improvements and bug fixes. Switching of meter weighting at run-time is now possible.
Improvements of the content authorship and right management system.
Implementation of arbitrary order 3D HOA encoding and decoding, with mode matching and ALLRAD decoder.
The internals of TASCAR are now available as a Open Access
journal paper:
Grimm, Giso; Luberadzka, Joanna; Hohmann,
Volker. A Toolbox for Rendering Virtual Acoustic
Environments in the Context of Audiology. Acta Acustica
united with Acustica, Volume 105, Number 3, May/June 2019,
pp. 566-578(13),
doi:10.3813/AAA.919337
With April 30, 2019, Ubuntu 14.04 LTS reaches end-of-life. Version 0.200 is the last version of TASCAR with support for Ubuntu 14.04.