|
The auditory scene (AS) is our mental picture of what is going on around us based on the sound we hear. For live sound, this scene is dominated by sight. Closing our eyes results in an AS that is a result of what we hear and partly what we remember from the visual scene. In sound reproduction, we want an AS illusion that we are listening to the live event. In fact, we are listening to two speakers in a small room. When we close our eyes, we want this reality to be replaced by an accurate AS illusion. With this testing, we hope to answer the question, Is this possible? Since the AS illusion depends heavily on having experienced the AS of live sound we require that the evaluator has recently attended live music (unamplified) in an appropriate acoustic venue. We expect that the listener was attentive and made notes of the live AS. Also, an attitude of "willing suspension of disbelief" is required. We are asking you to relax and let the illusion happen. Assuming it does, we ask you to critically judge its quality in 6 attributes by vertical marks cutting the scoring lines which go from bad to good. You are asked to listen to each of the segments at each of two listening locations to form an overall score for each of the attributes for each music or speech selections. Thus, for selection 1, you are asked to listen on one of the centerline seats and then in a seat to the side. At some point we ask that you mark a vertical line across the scoring line of each of the 6 attributes. You can play the segment as many times as you like and you can erase your mark and make a new one. Before going on to the next selection, write a small “1” above your mark on each scoring line. This will distinguish it from others marked “2,” “3.” Etc. when we decode your form. When listening, please do not limit what you hear by what you know to exist or what you can explain. We are listening for an illusion that we do not fully understand at this time. Also, please do not try to compare the Orion speakers to others as you imagine they would sound. Compare only to the original event AS as you believe you would perceive it if you were there. Use your experience and other cues to help define the original event. For instance, for LiT track 1, the wind, street noise and chirping birds tell us this recording was made outdoors. The voice should therefore sound dry and forward between the speakers. A reference speaker system known to be poor at generating the AS illusion is provided by the “A” switchbox option. “B” is the system under test. Written comments are encouraged. Please try to ignore the influence of: Purely spectral issues Loudness errors Emotional aspect of sitting in front of live musicians Detailed Instructions for Scoring Each AttributeIn General, a score of 0.0 means that, for a particular selection, the attribute is so distracting that the overall AS illusion is prevented. 5.0 means that any attribute deficiencies have a minor effect on the overall AS illusion. Scores over 5.0 mean the attribute adds to the AS illusion. Some problems come and go. If the problem is rare, giving a higher score should be considered. 1. Speakers Disappear Even if you can’t see them because of the opaque screen, all sound is coming from speakers. The desired illusion is that this is not heard as such. Score 0.0 or near if the sound seems to frequently emanate from either or both speakers. Score near 10.0 if there seems to be no relation of the sound to speaker locations. This illusion is recording dependant, so different selections may get widely different scores. 2. Local Acoustics Not Heard There is evidence that the acoustic signature of a room, such as our listening room, is heard only when first entering the room. After an adaptation period of a minute or so, the local acoustics should not define or constrict the size of the recorded illusion. Give a low score to any selection that gives an “in this room” impression. The local acoustic signature can be a “hollow” sound or a directional artifact, but is different from speakers not disappearing (attribute 1). Images Lateral Localization For most recordings the images should array evenly from left to right, sometimes extending the full angular width between the speakers (about 60 degrees). Distant images, especially, need not be sharply defined but all images should not shift. Give a low score for images split between two locations i.e. voice fundamental from one direction and sibilance from another. 4. Images Depth Localization A full range of distance illusion should be available from the speakers. This range will rarely be found in a single selection. If all images in a selection are perceived as medium distance, the score can be high if this is appropriate to the AS. The full test session shall contain selections covering the entire range of depth imaging. A single low score may weigh in heavily. 5. Ambience non-Localized Most recorded ambience is room reverberation, but it may include environment like wind and rain. Surrounding evenness is not always found at a live event, but directionality is never more than “frontal” or “behind.” Rotate your head to get a sense of excess directionality of ambient sound. 6. Freedom of Movement First, changing listening location from on-centerline to one seat off should not drastically change the image locations. Some change is okay. Second, at any listening location, small side-to-side movements or head rotation should not cause drastic image shifts. To Save Form right click on form and then click Save Image As
Send e-mail to SMWTMS: djcarlst@provide.net
Web Page Created by David Carlstrom.
|