The speaker characteristics of the sound-localization hoop setup at the department of Biophysics, Donders Institute for Brain, Cognition and Behaviour.
The speakers on the hoop are Visaton SC5.9 speakers (Haan, Germany). The manufacturer specifications can be found on the network drive: \\galba.azn.nl\mbaudit2\Utility\Specs_speakers. The speakers (SC5.9; Visaton) are mounted on a circular hoop with a radius of 1.2 m that can be rotatedalong a vertical axis. The hoop is described in detail in Bremen et al. (2010). A green LED is mounted at the center of each speaker and can serve as an independent visual stimulus. The intensity can be controlled by the stimulus computer. The spatial resolution of the hoop speakers in elevation is 2.5° ranging from −57.5 to +82.5°, while the spatial resolution in azimuth is better than 0.1° over the full range of 360°.
To determine the speaker characteristics in the hoop, we played a broadband Schroeder-sweep stimulus at several intensities (from [30:10:70] in the exp-file; see manual). Actually, twenty sweeps ( two sweeps are presented in figure 3A) were present in every stimulus, that was played at the 29 speakers in the frontal hoop. The sounds were measured with a Bruel & Kjaer (see figure 1 for recording setup).
The average sweep response and its power spectrum for every speaker and intensity is shown shown in figure23B and C, respectively.
All the data mentioned here is stored in PandA.
The power-spectra and their average for an intensity of 60 are shown in figure 3. The power spectrum is relatively flat and consistent over speakers, with a sharp decrease in power for frequencies less than 500 Hz, and no strong direction-dependent features.
The power of the recorded signal depends linearly on the intensity as given in the exp-file (for intensities between 30 and 70). This is shown in figure 4 for speaker 12 (straight ahead, averaged across frequency).
Based on the speaker characteristics in figure 5, we can build an equalizing filter.
After equalizing, the power spectrum has flattened, as shown in figure 6, although direction-dependent features due to reflections from the chair are still present.
Sounds are played over the speakers from the TDT real-time DA processor RP2.1. The setup is the same as shown above in figure 2. We played a 100 ms broadband Gaussian white noise, with no envelope, at several timings in the exp-file from speaker 12 (straight-ahead). The recorded timing and the duration is shown in figure 7. The actual stimulus-onset should depend on the timing in the exp-file (T), a 20-ms level ramp hardcoded in the RCO, and a delay from the speaker to the microphone (1.25 m / 340 m/s x 1000 ms/s = 3.7 ms).