Ambisonic Auralizer Tools VST User Guide

Transcription

1 Ambisonic Auralizer Tools VST User Guide Contents 1 Ambisonic Auralizer Tools VST Plugin installation B-Format Source Files Import audio files into host B-D Transcoder Basic routing configuration Plugin parameters bdtranscoder bdtranscoderstd PreAuralizer Plugin parameters Virtual Mic VST Plugin parameters B-to-UHJ Transcoder Plugin parameters UHJ-to-B Transcoder Plugin parameters Appendix A: Channel assignments in Wave Extensible files 18 1

2 1 Ambisonic Auralizer Tools VST Ambisonic Auralizer Tools VST is a VST 2.4 plugin bundle containing some useful tools to route and rotate audio signals in an Ambisonic 3D (or 2D) environment. The bundle consists of the following plugins: PreAuralizer bdtranscoderstd bdtranscoder b2uhjtranscoder uhj2btranscoder VirtualMic Use this bundle with any VST-compatible host software such as Cubase, Nuendo or Reaper. Learn about Ambisonics here: Download a short tutorial about Ambisonics here: pdf. 1.1 Plugin installation Copy the dll files (WINDOWS) or the bundles (MAC OSX) to your VST folder or configure your host program by setting up the plugins filepath. For example in Cubase/Nuendo: Devices Plug-in Information VST 2.x Plug-in Paths Add 2

3 Figure 1: Plug-in Information Window After adding the path in VST 2.x Plug-in Paths, update your plugin list by clicking on Update in Plug-in Information and get back to the main Cubase window (Cubase will update its plugin list whenever it is relaunched). 1.2 B-Format Source Files All the plugins in the bundle receive 4 INPUT channels and return up to 4 OUTPUT channels. You can produce your source files in 2 ways: 1. Use a 4-CHANNEL Wave Ex file containing the B-Format tracks (W,X,Y,Z): each track must be on a separate channel: W FL X FR Y BL Z BR (Front Left) (Front Right) (Back Left Surround Left) (Back Right Surround Right) Read how to import an audio file into your host program in Paragraph 1.3 later on. 2. Route 4 separate mono Wave files (W,X,Y,Z) to a QUADRO Group track and use the Group track as your source when you use the plugins. Surround routing can be accomplished by using the surround panner on each track (see Figure 2 for Cubase surround panner). 3

4 Figure 2: Cubase/Nuendo Surround Panner The 4 tracks should be panned as in Figure 3: Figure 3: Use the Surround Panner to pan the B-Format sources: W FL, X FR, Y BL, Z BR Read how to import an audio file into your host program in Paragraph 1.3 later on. In case the system is horizontal only, Z is not present (height information is not present). 1.3 Import audio files into host After creating a new project in your host program, import the B-Format signals W, X, Y, Z (or ONE single 4-channel Wave Ex file): File Import Audio File Select all B-Format audio files W,X,Y,Z (shift+click) and click Open. If needed, select the option Copy Files to Working Directory in the dialog window. Make sure you have selected the correct sampling frequency in your project (it has to be the same as that of the imported files) or, in case, convert your audio files to proper format. The dialog window shown in Figure 4 will open: 4

5 Figure 4: Dialog window: Choose Different Tracks! If you select Different Tracks, the audio files relative to signals W, X, Y, (Z), will be imported to different mono tracks. It is not necessary to import all B-Format files together (simultaneously). Alternately, you can import each single file to a single mono track. 2 B-D Transcoder B-D Transcoder transforms B-Format into D-Format in order to reproduce the Ambisonic recordings via a speaker array. The plugin comes in 2 versions bdtranscoderstd: horizontal regular standard arrays (angles are fixed) bdtranscoder: periphonic arrays (fully configurable angles in azimuth and elevation) bdtranscoderstd supplies 5 different standard array layouts defined by the azimuth angles (in degrees) below: four = [45, 135, 225, 315] five = [0, 30, 110, 250, 330] six = [30, 90, 150, 210, 270, 330] seven = [0, 30, 110, 135, 225, 250, 330] eight = [0, 45, 90, 135, 180, 225, 270, 315] Both plugins receive 4 INPUT channels and return 4 OUTPUT channels where only the 1st output channel is nonzero and it is intended to feed the i-th loudspeaker. 2.1 Basic routing configuration Follow the instructions below to route the signals properly before using the bdtranscoder and bdtranscoderstd: 1. We assume you have imported a 4-channel Wave Ex file. Select No bus in Cubase Inspector as the output of your audio track (Figure 5): 5

6 Figure 5: Cubase Inspector: select track output. 2. We assume you want to produce D-Format on 4 loudspeakers. Add 4 QUADRO Group tracks. Use Sends to send the audio track (BFormat Folk in the figures) to the 4 Groups (G1, G2, G3, G4) as in Figure 6: Figure 6: Use Sends to send the audio track to the groups. (If you want to produce D-Format on 5 loudspeakers, add 5 quadro groups and so on.) 3. Insert an instance of bdtranscoder or bdtranscoderstd in each group and configure each instance to produce the correct output signal (no. of speaker, selected speaker: read Paragraph 2.2). Do as Figure 7 to setup a 4 speaker standard layout. 6

7 Figure 7: Configure bdtranscoderstd. 4. Add one FX track for each speaker and set up the output of each group to the corresponding FX track (Figure 8): Figure 8: Configure the output of each Group track. 5. Route each FX track to the desired physical output (Figure 9): 7

8 Figure 9: Select the physical output for each FX track. How to configure physical inputs and outputs: press F4 to view and configure Cubase/Nuendo inputs and outputs in VST Connections (see Figure 10): 8

9 Figure 10: VST Connections 2.2 Plugin parameters bdtranscoder Figure 11: bdtranscoder plugin Periphonic This slider behaves as a switch (On/Off). If Periphonic is equal to 1, the plugin works in periphonic mode. If Periphonic is equal to 0, the plugin works in horizontal mode. N. Speakers: use this slider to select the number of speakers in the array (min: 4, max: 8). Azimuth (theta): select the azimuth angle θ i for the i-th loudspeaker. 9

10 Elevation (phi): select the elevation angle φ i for the i-th loudspeaker. Volume: Volume works simultaneously on all output channels. The plugin produces one output signal (feeding the i-th loudspeaker) as follows: S (SN3D) i = 1 L [W + X cos θ i cos ϕ i + Y sin θ i cos ϕ i + Z sin ϕ i ] PERIPHONIC S (N2D) i = 1 [W + X 2 cos θ i + Y ] 2 sin θ i HORIZONTAL L where θ i denotes the i-th azimuth angle, ϕ i denotes the i-th elevation angle and L is the number of the loudspeakers used in the array bdtranscoderstd Figure 12: bdtranscoderstd plugin N. Speakers: use this slider to select the number of speakers in the array (min: 4, max: 8). Sel. Speaker: select the output speaker. Volume: Volume works simultaneously on all output channels. 3 PreAuralizer This plugin offers some pre-processing tools to adjust the B-Format recording before transcoding from B-Format to D-Format or any other Ambisonic distribution format. It receives 4 INPUT channels (B-Format) and returns 4 OUTPUT channels (B-Format). Insert the plugin into your 4-channel audio track or quadro source group as in Figure 13: 10

11 3.1 Plugin parameters Figure 13: Open an instance of PreAuralizer Figure 14: PreAuralizer plugin Periphonic: this slider behaves as a switch (On/Off). If Periphonic is equal to 1, the plugin works in periphonic mode. If Periphonic is equal to 0, the plugin works in horizontal mode. Phase Eq IN: this slider behaves as a switch (On/Off). If Phase Eq IN is equal to 1, Gerzon s phase equalizer is inserted. If Phase Eq IN is equal to 0, the phase equalizer is bypassed. These filters are based on Gerzon s Phase Equalizer for Soundfield microphones. Gerzon s Phase Equalizer is recommended for Soundfield-like microphones, in which the capsule layout is symmetric with respect to the center of the array. The phase equalizer filters have the following transfer functions: 11

12 d F XY Z (s) = 6 F W (s) = 1 3 τ 2 s τs τ 2 s τs τ 2 s 2 + τs τ 2 s τs + 1 where τ = 340 and d is the capsule-to-center distance. Note: these filters are currently under revision. Caps-Distance: if Phase Eq IN is equal to 1, the slider named Caps- Distance allows to set up the distance of the capsules from the center of the microphone array. Shelf IN: this slider behaves as a switch (On/Off). If Shelf IN is equal to 1, two classic shelf filters are inserted (high shelf and low shelf). If Shelf IN is equal to 0, the shelf filters are bypassed. LF HF Horizontal W 0 db (1.76 db) Horizontal XY 0 db 0.86 (-1.25 db) Periphonic W 0 db (3.01 db) Periphonic XYZ (1.76 db) 0 db Table 1: High Shelf and Low Shelf gains Dist Comp IN: this slider behaves as a switch (On/Off). If Dist Comp IN is equal to 1, a distance compensation highpass filter is inserted. If Dist Comp IN is equal to 0, the filter is bypassed. Array Distance: if Dist Comp IN is equal to 1, the slider named Array Distance allows to set up the distance of the loudspeaker array from the listener (placed in the center of the array). The more you decrease the distance the higher the cutoff frequency gets. N. Speakers: use this slider to select the number of speakers in the array (min: 4, max: 8). Volume: Volume works simultaneously on all output channels. 12

13 4 Virtual Mic VST This plugin supplies 3D virtual rotations, zooming effects and microphone pattern shaping. It receives 4 INPUT channels (B-Format) and returns 4 OUTPUT channels (B-Format). Insert the plugin into your 4-channel audio track or quadro source group as in Figure 15: 4.1 Plugin parameters Figure 15: Open an instance of VirtualMic Figure 16: VirtualMic plugin Yaw Pitch Roll Use these sliders (Yaw, Pitch, Roll) to rotate the virtual microphone. Set up 13

14 the rotation angles by varying the values of the sliders. Yaw: rotation about axis z. Pitch: rotation about axis y. Roll: rotation about axis x. Zoom This parameter controls dominance. Its range goes from -1 to 1 and it represents Malham s parameter µ in Malham s dominance matrix. When µ = 0 the zoom effect is inactive. Malham s dominance matrix is defined as L µ = 1 µ 0 0 µ µ µ 2 Set Pattern In This slider behaves as a switch (On/Off). If Set Pattern is equal to 0, the shaping of the virtual microphone s polar pattern is deactivated and the outputs produced by the plugin correspond to the B-Format signals after rotation. If Set Pattern is equal to 1, then Select Pattern allows to choose the desired polar pattern. The plugin supplies only one output (the first: Left Front) which is the output of the virtual microphone. Select Pattern The parameter Select Pattern selects the microphone directivity factor D. The options include: D = 0 Omnidirectional D = 0.5 Sub-cardioid D = 1 Cardioid D = 1.5 Hypercardioid D = 2 Figure of 8 Table 2: Directivity Factor Volume Volume works simultaneously on all output channels. 5 B-to-UHJ Transcoder This plugin allows transcoding from B-Format (W,X,Y,Z) to UHJ-Format (Σ,, Q, T). It receives 4 INPUT channels (B-Format) and returns 4 OUTPUT channels (UHJ-Format). Insert the plugin into your 4-channel audio track or quadro source group as in Figure 17: 14

15 Figure 17: Open an instance of b2uhjtranscoder 5.1 Plugin parameters Figure 18: b2uhjtranscoder plugin Select L/R This slider behaves as a switch (On/Off). If Select L/R is equal to 1, the plugin produces 2 outputs (LEFT FRONT and RIGHT FRONT) corresponding to signals Left and Right derived from UHJ. If Select L/R is equal to 0, the plugin outputs are the UHJ-Format signals Σ,, T, Q: Out 1 Σ Out 2 Out 3 T Out 4 Q Volume Volume works simultaneously on all output channels. 6 UHJ-to-B Transcoder This plugin allows transcoding from UHJ-Format (Σ,, Q, T) to B-Format (W,X,Y,Z). It receives 4 INPUT channels (B-Format) and returns 4 OUTPUT channels (UHJ-Format). 15

16 Insert the plugin into your 4-channel audio track or quadro source group as in Figure 19: Figure 19: Open an instance of uhj2btranscoder The plugin input is UHJ-Format. As in B-Format, you can use a 4-channel Wave Ex file or a QUADRO group as the source. Important: the order of the source input tracks must be as follows: In 1 Σ In 2 In 3 T In 4 Q 6.1 Plugin parameters Figure 20: uhj2btranscoder plugin Select L/R This slider behaves as a switch (On/Off). If Select L/R is equal to 1, the plugin produces 2 outputs (LEFT FRONT and RIGHT FRONT) corresponding to signals Left and Right derived from UHJ. If Select L/R is equal to 0, the plugin outputs are the B-Format signals: W,X,Y,Z: 16

17 Out 1 W Out 2 X Out 3 Y Out 4 Z Volume Volume works simultaneously on all output channels. 17

18 Appendix A: Channel assignments in Wave Extensible files The default assignment order for interleaved channels on a Wave Extensible audio file is shown in Table A.1: 1 SPEAKER FRONT LEFT FL 0x1 2 SPEAKER FRONT RIGHT FR 0x2 3 SPEAKER FRONT CENTER FC 0x4 4 SPEAKER LOW FREQUENCY LF 0x8 5 SPEAKER BACK LEFT BL 0x10 6 SPEAKER BACK RIGHT BR 0x20 7 SPEAKER FRONT LEFT OF CENTER FLC 0x40 8 SPEAKER FRONT RIGHT OF CENTER FRC 0x80 9 SPEAKER BACK CENTER BC 0x SPEAKER SIDE LEFT SL 0x SPEAKER SIDE RIGHT SR 0x SPEAKER TOP CENTER TC 0x SPEAKER TOP FRONT LEFT TFL 0x SPEAKER TOP FRONT CENTER TFC 0x SPEAKER TOP FRONT RIGHT TFR 0x SPEAKER TOP BACK LEFT TBL 0x SPEAKER TOP BACK CENTER TBC 0x SPEAKER TOP BACK RIGHT TBR 0x20000 Table 3: Wave Extensible: dwchannelmask assignments The Format Chunk of a Wave Extensible file (the format is indicated in 18

19 the Format Chunk as WAVE FORMAT EXTENSIBLE) includes a 4-byte field named dwchannelmask. This field has the purpose to assign a channel in the Wave file to a precise audio output (hardware or software, if anything like a virtual patchbay or internal routing system is present). Pay attention to the field named nchannels in the Format Chunk during the assignments. This field indicates the number of channels actually present. If its value is inferior to the number of bits set up in dwchannelmask, then the extra bits (most significant) of dwchannelmask are ignored. Else, if its value is greater, then the remaining channels are not assigned to any particular output. An audio device should route the remaining channels to unused outputs. Figure 21: Wave Format Extensible dwchannelmask assignments: normal positions (top positions) When dwchannelmask is set to 0, the audio device will play the first channel via output 1, the second via output 2, etc: devices like digital mixers or multitrack recorders accept audio formats regardless of any precise indication about position and follow the order of the channels as they are recorded on the file. The value 0xFFFFFFFF in dwchannelmask indicates that all possible channel configurations are supported. 19