CATT-Acoustic v8

1 CATT-Acoustic v8. Room Acoustics Prediction and Desktop Auralization TM. tim tim asu ea easu ces a so utio lutio ing n k rs s rati atio al m rica cal m er p ion onv conv -trac tou ard on o ad e ( e (T res sur res sin un n dc ,L m yc vis ever ver ume num ume spe raliz -line tenc he ). DT gra ivit th T. ro vi ol y s S. ep of of of ots o s of r lou op a gh o ow- n wi ica -30 , E ho ct RA. g ss 5, T -80 , ec ire The OpenGL-based CATT 3D-viewer offers c io la F, d enhanced 3D-viewing with surface color- led rien , co lat ati ati s/p ma ral esk ro ime at , ow pe as lcu tim tim ph lor au t d lkth l-t aliz coding (material, absorption factor, scattering .). ri ak at l ff etc coefficient, specularity, diffusitivity), variable Kn Ex Ide Ca Es Es Gra Co Bin Fas Wa Rea Aur es lighting, viewpoint lists, animated viewpoint ine transitions, walkthrough with auto-playing u u l, S ).

2 C. ab auralized WAV-files and many viewing options. (. e lm The viewer is a small stand-alone application d t cla -1. that is ideal for customer presentations. e n o ion be ber ric on n p f l re n ge ce nc ion on on r La DS ke ts, P. CA. TT. -A. co us tic Yo u CATT-AcousticTM v8 is a seven-module integrated application for 32-bit Windows . It combines numerical prediction, multiple source addition, convolution, auralization, sequence (batch). processing, source directivity and surface property library handling. In addition, the stand-alone OpenGL-based CATT 3D-viewer can be used for enhanced 3D-viewing and customer presentations. Copyright CATT 2003 All products mentioned are trademarks of their respective owners The Prediction module utilizes the unique Randomized Tail-corrected Cone-tracing (RTC-II), the Image Source Model (ISM) as well as Ray- tracing.)

3 The module predicts room acoustical numerical measures, and creates data files for multiple source addition and for further post-processing for auralization. Based on results from the Prediction module, the Multiple source addition module creates sum echograms by addition of individual source echograms where source directivity, aim, eq, background noise and delay can be varied without need for a full recalculation. Source changes can also be applied to individual source echograms. The module optionally creates new data for multiple source auralization. All steps from prediction over binaural post- Octave-band echogram reflections, processing, filtering to convolution and WAV- created by the prediction module, are via file calibration can be performed in "batch".

4 DSP procedures and HRTFs transformed using the Sequence processing module. to binaural reflections creating a binaural Sequence merge, re-arrange, loops and string impulse response to be convolved with variables give flexibility and power. anechoically recorded material - auralization. The Post-processing module offers many other post-processing options (mono, stereo, B-format, 5-channel), multiple source auralization, fast convolution, walkthrough convolution, crosstalk cancellation, headphone equalization, and an assortment of file format conversions, scaling, filtering, samplerate conversion and calibration utilities as well as a WAV-file player with A/B and a playlist. 50 dB axis DI= dB. 500 Hz Up Lt Back Front Rt Dn Directivity balloon, 10 -format (sampled every 5 ).

5 D-50 [%] 1 kHz C-80 [dB] 1 kHz 60 4. 3. 50. A0 A0. Source directivty is either imported from measured 10 full-space 2. B0 40 B0. data-files, run-time simulated via a DLL-interface (DDI, with 1. array modeling including handling of the near-field) or 30. 0. interpolated from simple 15 polar data. 20 -1. SPL [dB] 1 kHz D-50 [%] 1 kHz 78 60. 77. A0. 50. 76. B0 75 40. 74. 30. 73. 72 20. Audience area color mapping (palette can be selected). dB Complete echogram dB Early echogram 80 80. 70 70. 60 ( ) 60. ( ). 50 50. 40 40 4ms Filter 30 30. Named surface properties are either managed by the Surface 20 20. 10 10. properties module or entered directly into geometry-files. Each 0 500 1000 1500 ms 0 25 50 75 100 125 150 175 ms 23 48 73 98 123 148 173 198. user can have his own private library selected in Prefs.

6 EDT s dB. 80. Early directional echograms T-15 s T-30 s 70. Graphical output can 60 Up D-50 %. be transferred over C-80 dB. 50. Lt Rt LEF1 % 05 40. the clipboard (Meta LEF2 % A0 30. Dn Ts ms OMNI. and Bitmap) or be ex- DI = 0 dB. 20. SPL dB dB at 1 m 10. ported in EMF, BMP G-10 dB 1 kHz 5m 0 25 50 75 100 125 150 175. 23 48 73 98 123 148 173 198. ms and EPSF formats. Full calculation echograms (RTC-II). Plot-files can be dB A0 viewed in sequence 70. interactive or auto- 60. matic and with auto- 50. playing WAV-files attached to plots. A stand-alone plot-file viewer 40. 30. can be used for cus- 20. tomer presentations. 0 50 100 150 ms The stand-alone dB B0 70. OpenGL-based 3D- 60. viewer offers en- 50. hanced viewing with 40. variable lighting and 30. many options such as 20.

7 0 50 100 150 ms surface color coding. Source addition: individual + total echograms % LEFT EAR dB H_A0_05_1K Receiver angles 100 70 30. 60. 50. 50. 0 40. 30. -50. 20. -100 10. 0 50 100 150 200 ms 0 50 100 150 ms 23 73 123 173 223 23 73 123 173 Ver Hor % RIGHT EAR S-300-102-R. Source angles 100. 50. 31. 21 32. 0 11. 01 12 22 33. A0 02 13 23 34. -50 03 14 24 35. 04 15 25. 05. -100. 0 50 100 150 200 ms 23 73 123 173 223 Ver Hor Synthesized binaural room impulse response Interactive specular reflection path analysis (ISM). See for many examples of real-world hall models as well as all graphical output from CATT-Acoustic v8. General CATT-Acoustic v8 employs a unique and general Room Early Decay Time (EDT) T-15 T-30 Eyring Prediction prediction method introduced with v7 and baptized Acoustic and Sabine (interactive) SPL (also dBA) Strength Randomized Tail-corrected Cone-tracing (RTC-II), Methods that combines the best features of the Image Source Measures (G) Speech Transmission Index (STI/RASTI).

8 Definition (D-50/C-50) Clarity (C-80) Lateral Model (ISM), Cone-tracing and Ray-tracing. All Energy Fraction (LF, LFC) Center of Gravity (Ts) . methods take into account frequency dependent diffusion (the ISM for Inter Aural Cross-correlation Coefficient (IACC, after binaural post- first order reflections): walls, or parts of walls, can be assigned frequency processing) Mean absorption factor and mean free path Measures can dependent absorption, surface and edge scattering, and transparency be normalized with expected values Spreadsheet export. coefficients. Due to the frequency dependence of diffuse reflection, separate ray/cone-tracing is performed for each octave. With RTC-II, direct sound and 1 st+2 nd order specular reflections are handled Graphics General: 3D models with selectable shading palettes deterministically using the ISM while 1st order diffuse reflection is (natural, parallel, and camera with variable lens) or handled by direct reflection from diffusing surfaces.

9 Higher order diffuse actual surface colors x-y, x-z, y-z views Interactive source information reflections are handled by randomly distributing rays that hit diffusing including -3/-6/-9 dB contours Reflector coverage contours Animated surfaces. Dedicated ISM can be used for qualitative early reflection calculation of rays and echogram. ISM: Interactive reflection tracing . analaysis while standard Ray-tracing is used for audience area color Echograms (dB-scale backward and forward integrated, linear, ear- mapping. Predictions are made for up to 260 sources, 100 receivers and integrated with Sone scale) Image source 3D space. RTC-II: Echograms 8 octave-bands (125 Hz - 16 kHz, the 8 kHz and 16 kHz bands are (full-length, detailed, detailed directional, smoothed, backward and extrapolated if no data is available).)

10 In a round-robin test on computer forward integrated) Vectorgrams Sound roses Absorption statistics modelling software, only three out of sixteen software were judged to Reflection path statistics Interactive wall hit statistics Reverberation give reliable and useful results1. Of these three CATT-Acoustic (v5) decays Room acoustic measures overview and trend RASTI STI . had the best results for five of the eight measures calculated, methods Octave-band Transmission Index Animated rays and echograms. Ray- employed by v7 and v8 are of a more general nature and give overall tracing: Direct sound, Early part SPL and LF mapping Color mapping even better results. of most measures (T', SPL, G, STI, RASTI, D-50/C-50, C-80, LF, Ts, source delay, closest source group) 2D and/or 3D maps with selectable scales/palettes Animated rays and maps.