On Tues. April 13th, 2010 the Boston Section of the Audio Engineering Society featured David Griesinger speaking on "The Relationship Between Audience Engagement and Our Ability to Perceive the Pitch, Timbre, Azimuth and Envelopment of Multiple Sources". The talk was broken up into three parts: Physics, Psychology and Acoustics.
Dr. Griesinger has spent many years trying to figure out how to actually measure what you hear. Sound coming from different directions interferes with each other causing major level differences between the two ears. He addressed some misconceptions of how the basilar membrane receives information and the importance of pitch in determining the distance of a sound. He showed that the human ear has evolved to analyze fluctuations or modulations in the amplitude of the basilar membrane motioned at frequencies above 1000Hz. He presented the design of a recently discovered neural mechanism for human hearing that explains our ability to separately perceive the pitch, timbre, azimuth and distance of individual instruments in a string quartet, or a number of humans speaking at once. This ability depends on the phase coherence of harmonics in the frequency range of the vocal formants. When the phases of these harmonics are not scrambled by reflections, performers are perceived as "close" and demand the full attention of the listener. All these perceptions depend on the ears' ability to perceive the direct sound from the source.
Objective data was presented obtained from analyzing harmonic coherence in binaural recordings of live music, along with subjective assessments of the clarity of sound. The high degree of correlation between these two raises the prospect that engagement in occupied concert halls can be measured unobtrusively. He discussed design features of halls that combine high clarity of direct sound along with sufficient late reverberation.
Examples of a voice counting to ten were played with the same loudness and frequency content but with different harmonic content. The phase relationships of harmonics from a complex tone contain information about the sound source that created them. These phase relationships are scrambled by early reflections. Reverb is often louder from the front than the back of the room. Dr. Griesinger expressed his belief that the psychological impact when direct sound is present is completely different than when it is not. Delaying the reverb by 20ms created more clarity with the speech example. It became more enveloping and louder when the direct sound was added. He proposed that early reflections in the time range of 10-100ms reduce clarity, envelopment and engagement and demonstrated this with his listening examples.
David Griesinger is a physicist and recording engineer. He designed the algorithms and hardware architecture of the Lexicon reverberation units, the Logic 7 matrix decoding system, and has recorded hundreds of (mostly classical) music CDs. He lectures on recording technique, microphones, headphones, and the interaction between loudspeakers and rooms. His current interest is largely in the field of concert and opera acoustics. He has been striving to understand the physics and physiology of human hearing, in order to better design spaces for music listening. He lives in Cambridge MA, where he enjoys his family, singing and HD music videography.
