Portal de Periódicos da CAPES

A Survey Study of In-Situ Stereo and Multi-Channel Monitoring Conditions

2001; Audio Engineering Society; Linguagem: Inglês

1549-4950

Aki Mäkivirta, Christophe Anet,

Advanced Adaptive Filtering Techniques

The in-situ responses of a total of 372 loudspeakers in 164 professional monitoring rooms around the world have been measured after acoustical calibration. All measured rooms have been equipped with factory calibrated three way monitors and acoustically calibrated with standardized apparatus. The results provide a thorough understanding of typical monitoring conditions for stereo and multi-channel rooms, distribution in room parameters and quality of reproduced audio. Results are compared to current standards and recommendations. INTRODUCTION The audio quality of monitoring determines the resulting quality of the recording process. The monitor loudspeakers strongly effect the aesthetic decisions of artists and engineers producing audio material. The installation of the speakers and the monitoring room itself has typically an even stronger influence, and an increasing number of standards and recommendations have been issued to ensure consistency and high quality of the monitoring process both with and without an accompanying picture [1-5]. These give recommendations of aspects of in-situ reproduction quality such as • magnitude response at the listening location • dynamic range and noise level • time-of-flight difference • idealized layout configuration • nominal listening level • reverberation and reflection in the room • quality of the reproduction system METHODS All rooms included in the present study are equipped with a factory calibrated 3-way main monitors, produced by one manufacturer. Measurement Apparatus and Method The measurement method used an MLS sequence of period 16383 samples (217ms) at a sampling rate of 75.47kHz. An impulse response of length 16383 samples was stored for each loudspeaker. The measurement apparatus was laboratory calibrated to a reference measurement system before and after the measurements. The microphone was calibrated by producing a separate calibration file that was used during analysis to equalize the magnitude response calculated from an impulse response. To ensure high consistency of the measurement process one person used the measurement appa1 Genelec OY, Iisalmi, Finland 2 MLSSA measurement system, DRA Laboratories. 3 Neutrik Type 3382. MAKIVIRTA AND ANET IN-SITU MONITORING CONDITIONS AES 111 CONVENTION 2 ratus throughout the study to perform all on-site calibrations and measurements. Detection of impulse response start The onset of an impulse response is determined as 10 samples before the impulse response power estimate exceeds 3% (–30dB) of its peak value. This corresponds closely to an impulse response start definition given in standards of room reverberation measurement [6]. Signal-to-Noise Ratio and Listening Distance The stored impulse response contains the natural time-of-flight before the impulse begins. This section in the recorded signal represents the total system plus room noise level, and it is used as an indication of the quality of measurement (signal-to-noise ratio) and measurement distance. The signal-to-noise ratio D of a measurement is calculated as a ratio of the recorded impulse response data a(t) peak value to the noise level rms value within the time-of-flight period [t0, t1] before the onset of the actual impulse response