Header image  
by Ralph Glasgal
 
line decor
Home Tutorials Tech
Papers
Kudos and
Pictures
Demos Bio Free Ambio
book
Glossary The Home
Concert Hall
PC/Mobile
Applications
Rec Engineers
Corner
FAQ/Forum Links Contact us
line decor
  Ambiophonics Terminology
The language of stereo-and-5.1-compatible loudspeaker-binaural
Robert E. (Robin) Miller III, AES SMPTE BSEE, ©2009
Filmaker Technology . www.filmaker.com

A critical method of sound reproduction for entertainment (music, movies, games, rides), Ambiophonics has its own lexicon. This paper provides useful definitions of unique Ambiophonics terms deemed essential for understanding by audiophiles and audio engineers.
In sum, the root name Ambiophonics is given to several related audio technologieswhich improves imaging/localization, spatiality (envelopment, immersion), and "you are there" of sound reproduction of music, movies, and games. Championed by Glasgal, Farina, and the author, Ambiophonics has been in development since 1988, with levels of perceived “reality” ranging from front-only presentation to life-like 3D, including
:
  1. Reproduction of 2-channel recordings intended for stereo (60° front-only) that extends the perceived reproduced front stage width to 120° or more (with or without the addition of convolved ambience) – known as Ambiophonics (Fig1);
  2. Enveloping 360° (horizontal plane) reproduction of multi-channel surround sound recordings, e.g. ITU-R775 standard 5.1 or 4.0 (implying directly-recorded ambience or direct sources around sides/back) – known as PanAmbio (Fig2);
  3. Immersive periphonic 3D reproduction (with height, up to the full sphere of human hearing, directly recorded) using special 6-channel recordings – known as PerAmbio or TriAmbio.
Note: Not related to “Ambiophony,” introduced in 1960 by Parkin as a method for electronically simulating acoustic reflections in concert halls that are lacking in reverberation, especially for symphonic music.
Ambiophonics diagram
Fig 1: Ambiophonics’ speakers FL & FR convey a stage equaling the original recording angle extending to virtual FL & FR, uncolored center voices, and maximal listener envelopment (LEV), reaching within the regions shown, all in contrast to that of conventional stereo speakers L & R.
PanAmbiophonics
Fig 2: PanAmbio surround adds back speakers BL & BR, imaging as BL & BR – four “speakers” (double 5.1/7.1’s two) within regions where listener envelopment (LEV) is maximum. (Play 5.1 in PanAmbio by setting the player to “no center” to mix the C-channel to FL and FR.)

 

The Ambiophonics “family”…

Ambiophonics - (Fig1) loudspeaker-binaural reproduction of 2-channel LP/CD/MP3/SACD, etc., which processes a stereo (2.0) pair of signals and presents them to a centered listener by an Ambiodipole (closely-spaced speaker pair) in front. Optionally, image width may be augmented by duplicating the signal pair in a rear Ambiodipole. Additionally, to restore a natural level of ambience that has been deliberately reduced in commercial stereo releases, two or more Ambiostats (surround speakers) driven by hall impulse response convolvers DSP provide 2D or even 3D surround reverberation based on two media channels (including most legacy stereo recordings on CD, LP, games or broadcasts).

Ambio - colloquial usage, as stereo is for stereophonics.

Panorambiophonics - (PanAmbio 4.0/4.1, see Fig2 ) Uses independent front and back Ambiodipole pairs of speakers to reproduce 360° sound fields from 4.0/5.1 media. Records and replays two Ambiophonics stages – a front stage typically of direct and frontal reflected sounds, and a stage behind the listener for reflected and ambience sounds, along with any direct sources around the sides and in back. In contrast to 5.1, PanAmbio is isotropic (equal in all directions) in the horizontal plane. It renders accurate 360° localization ±5° even at the extreme sides and avoids localizing to 5.1’s speakers at 110°.

Perambiophonics - (PerAmbio 6.0/6.1) Full 3D reproduction which is spherically isotropic, as with live, natural human hearing. TriAmbio adds a third, elevated Ambiodipole for reproducing height using special 6.0/6.1 recordings.

Ambiophonics’ hardware…

Ambiopole - An optional loudspeaker optimized for use with digital crosstalk cancellation processing, having controlled dispersion and phase-consistent performance in any crossover regions.

Ambiodipole - A closely-spaced (~20°) pair of speaker Ambiopoles placed in front of and optionally behind listeners.

Ambiophone - two-channel main microphone used to make stereo-compatible two channel recordings for critical Ambiophonic reproduction according to psychoacoustic principles.

Panambiophone - main 4-channel microphone array used to make both stereo-compatible two channel recordings with ambience control, or 4.0 (5.1-compatible media) PanAmbio surround recordings.*

Ambiostats - surround speakers used to radiate concert- hall reverberation derived by convolution of hall impulse responses.

Ambiovolver – PC with DSP software for both crosstalk cancellation and hall ambience convolution. Two instances of crosstalk cancellation are required for panambio surround. Fig 2: PanAmbio surround adds back speakers BL & BR, imaging as BL & BR – four “speakers” (double 5.1/7.1’s two) within regions where listener envelopment (LEV) is maximum. (Play 5.1 in PanAmbio by setting the player to “no center” to mix the C-channel to FL and FR.)

Related audio formats & scientific terminology…

Stereo, stereophonic 2-channel audio reproduction attributable to work in the 1930s at Bell Labs and EMI.

“5.1” international surround standard ITU-R775 for 5 to 7 speaker channels plus Low Frequency Effect (LFE, aka “0.1”) for home theaters, evolved from multi-channel cinema sound since the 1939 movie “Fantasia.” PanAmbio is 5.1-compatible; PerAmbio and TriAmbio are 6.1/7.1-compatible.

Binaural hearing - (normal, everyday hearing mechanism) – this term is also often used to describe lifelike reproduction recorded with a surrogate “dummy head” with microphones at its ear positions, then played back using headphones. When reproduced directly into the listener’s ear canals such as with intra-aural ear receivers (“ear-buds”), the dummy head has generic pinna that might not work with an individual listener. Pinna-less microphones such as the PanAmbiophone, reproduced over loudspeakers, involve listeners’ own pinna, and so satisfy life-like perception for all.

ILD, ITD interaural level & time (phase) difference in binaural ear signals from a sound source, registered by auditory sub-brain structures, which contribute to the perceptual qualities of localization, spatiality, timbre. About equally splitting the audible spectrum (but with much overlay), ITD is the controlling mechanism for sounds with frequencies in the five octaves below 700Hz and ILD for the five octaves above 700Hz (depends on listener’s head size).

Localization (imaging) the ability of humans (using IDD, ILD and pinna binaural cues) to identify within a sphere of possible arrival angles the direction of origin of sounds, both those that arrive along a line from the source (direct sounds) or of temporally (echo) distinguishable reflected (indirect) sounds. Acuity within an ellipse in front of the listener is about ±1° horizontally and ±10° vertically.

LEV (listener envelopment) a sense of being surrounded in 2-dimensions (or better yet immersed in the full-sphere 3D of human hearing); essential to human perception of the space containing the listener and sources of sound. This quality of space is largely missing in conventional stereo reproduction, but is much improved using Ambiophonics, where the ambience sounds especially may be localized within direction-dependent regions of maximum LEV (see Fig1). PanAmbio surround is capable of regenerating more LEV than 5.1 surround (see Fig2).

Spatial resolution the degree, in contrast to either front-only stereo or to 5.1, by which 2D ambio, panambio surround sound, or full-sphere 3D preserve individual voices amid the ensemble and allow listeners to perceive separation of voices and events, as in live hearing. Heroics by stereo mixers to maintain audible “layers” (singer discernable from instruments, etc.) are needed far less given the greater spatial resolution of surround sound.

Reverberation Physically, the acoustical impulse response of a reflective room or space. The more reflective or larger the space, the longer lasting the reverberation.

“Reverb” jargon connoting artificial reverberation, sometimes acceptable, but often phony-sounding. Realistic results are either directly recorded in a reverberant space, or carefully convolved using a 2D or 3D hall impulse response.

Impulse; Impulse Response (IR) momentary non-zero event (e.g. hand-clap, gun-shot, balloon burst, or electronic "Dirac” pulse); which characterizes electro-acoustic systems – its Impulse Response (“IR”, e.g. reverberation). Each sample of a digital recording can be considered a Dirac impulse, which spawns its own IR by convolution (e.g. hall ambience by DSP).

Convolution using a digital recording and applying DSP signal processing to “imprint” an impulse response on a source signal. For example, multiplying and adding to the samples of an instrument or voice the acoustical impulse response of a hall as though the source had been recorded in that hall.

DSP digital signal processing. Used in Ambiophonics for crosstalk cancellation and optional ambience convolution. DSP for “XTC” include single delayed inversion (Wareing), inverse impulse response (Farina), band-splitting (Choueiri), and RACE (Glasgal). DSP for generating ambience convolves 2channel stereo recordings with 2D or 3D hall impulse responses.

Pinna the outer ear, whose fleshy convolutions cause peaks and dips in the frequency response depending upon the 3D angle of arrival of a sound. Together with ear signal difference ILD and ITD, pinna filtering encodes direction for the conscious brain to perceive localization. Pinna contradiction results when reproduction systems (notably conventional stereo and 5.1 surround) create phantom images the brain interprets as coming from a fuzzily-perceived direction despite being pinnaencoded as coming from their actual direction, e.g. a speaker.

Phantom image localization of a sound, the perceived direction of which is not real but the result of “conspiring” cues, such as two loudspeakers playing the same signal, perceived as coming from a point between the speakers. In conventional stereo reproduction, the common practice of “panning” a single microphone of a soloist equally to both channels/speakers is determined by the brain as coming from the center, although not convincingly (due to pinna confusion) and exhibiting a raspy comb-filtering. Phantom images to one side of center are drawn in non-linear fashion toward one speaker or the other.

Comb filtering distortion of a sound’s spectral character (frequency response) due to wave cancellation alternating with reinforcement when the identical signal arrives simultaneously from two points (e.g. speakers). Comb filtering due to crosstalk in stereo affects the frequency response at each ear of important central voices (e.g. soloist, dialogue), and unfortunately causes some mixing engineers to compensate by boosting equalization of the isolated central sound around 2kHz that permanently scars the recording for headphone or Ambiophonic replay. (The center “C” channel/loudspeaker solves this condition for 5.1 surround.)

Crosstalk unintended presence of reproduced sound from any speaker to the ear on the opposite side. Avoided by using headphones, speaker crosstalk at either ear is accompanied, for important central voices, by duplicate sounds, delayed in time, that interfere with earlier sound from the speaker on the same side, resulting in raspy-sounding comb-filtering and pinna-confusion in localizing the sound. Crosstalk is the principal impetus for Ambiophonics solutions.

Crosstalk cancellation (XTC, CTX) Ambiophonic solutions address the problem of crosstalk: if it can’t be avoided it in the first place, cancel it. Approaches to XTC all attempt to segregate speaker sounds to the intended ear. Methods over the years include a physical barrier projecting from the listener’s nose, adding speakers recessed and therefore delayed in time, and digital signal processing to introduce cancellation signals into the audio path, as with Ambiophonics’ RACE.

RACE (Recursive Ambiophonic Crosstalk Eliminator) method developed for creating (using DSP) properly timed and attenuated recursive crosstalk cancellation signals. Crosstalk is cancelled acoustically at listeners’ ears.