DOI: 10.14489/vkit.2015.03.pp.003-008

Лаврова Г. А.
ОЦЕНКА ХАРАКТЕРИСТИК ТЕХНОЛОГИИ 3D-АУДИО, РЕАЛИЗОВАННОЙ С УЧЕТОМ ПОВОРОТА ГОЛОВЫ ИСПЫТУЕМОГО
(с. 3-8)

Аннотация. Рассмотрена одна из возможностей улучшения интерфейса кабины самолета – применение технологии пространственного звука (3D-аудио). Технология реализуется путем подачи в наушники пилота обработанных звуковых сигналов. Используются методы цифровой фильтрации: перекрытия с суммированием и алгоритм вычисления свертки на основе быстрого преобразования Фурье. В качестве импульсных характеристик взяты данные из общедоступной библиотеки CIPIC HRTF Database. Предложено использовать микромеханическую бесплатформенную инерциальную навигационную систему для корректного задания положения виртуального источника звука. Представлены результаты экспериментов, проведенных с использованием программного обеспечения MATLAB и VisualStudio, по оценке характеристик точности локализации направления звука при модулировании широкополосного шума.

Ключевые слова: пространственные аудиоподсказки; антропозависимая передаточная функция; антропо-зависимая импульсная характеристика; звуковой интерфейс.

Lavrova G. А.
ESTIMATION OF CHARACTERISTICS OF 3D-AUDIO TECHNOLOGY, WHICH WAS REALIZED, TAKING INTO CONSIDERATION LISTENER’S POSITION OF THE HEAD
(pp. 3-8)

Abstract. The paper is dedicated to realization of 3D-audio technology through headphones and its further implementation in the cockpit. Spatial audio-cues, which can be generated by using this technology, will improve the audio interface of the aircraft’s cabin. To realize the technology it is necessary to provide processed audio signals to supplying pilot’s headphones. For sound signal’s processing finite impulse response filtering FFT-based overlap-add method is used. Impulse characteristics are taken from the public data library CIPIC HRTF Database. Tracking position of the listener's head is a necessary condition for valid implementation of the technology – to set correctly the position of the virtual sound source. For these purposes micromechanical inertial system device is proposed to use. By means of MATLAB and Visual Studio the software was created. It performs the necessary algorithms for audio processing and tracking the relative position of the listener's head and the virtual sound source. Using the software and micromechanical inertial system device an experiment was carried out. Modulated wideband noise was used as sound signal. Experimental group consisted of 7 operators with no obvious hearing defects. The main purpose of the experiment was to estimate accuracy characteristics, typical for listener’s capability to determine the direction of the virtual sound source. During the experiment processed audio signals was provided to subject’s headphones. Operator turned his head to “look” at virtual sound source as he heard it’s location. In that moment information about his head position was determined by information from the sensor. Difference between established and determined angle position was fixed as deviation (error) separately for horizontal (azimuth angle) and vertical (elevation angle) directions. The mean of azimuth angle’s deviation absolute value was 7.3 degrees. Similarly the mean for elevation angle was 26 degrees. These mean estimates were calculated for all the operators and positions in the experiment. The article also compares the results mentioned above with earlier experiment, which was carried out without head position sensor. In that case listener’s head position was fixed, and listener just expressed verbally his suggestions about angular position of virtual sound source. Though these experiments were carried out with different conditions, their results were approximately compared. Conclusion of this comparison is that accuracy characteristics of 3D-audio technology, which was realized with tracking listener’s head position, are approximately two times better.

Keywords: Spatial audio-point; Head related transfer function; Head related impulse response; Sound interface.