Notes
Slide Show
Outline
1
Rise Above the Jet Noise
  • AVX  Adaptive Encoding


  • Jeremy Hoenack,CAS, MPSE
  • Sound Trax Studios
  • jhoenack@avxsound.com
2
AVX Inflight Sound  Overview 1
  • Subtle dialog is clear and easy to understand
  • Underscore and background effects that are normally lost are crisp and clear
  • Film sound-track masks jet noise instead of reverse
  • No high sound levels, no ear pain, no need to constantly adjust volume, natural sounding
  • More immersion in film – more positive flight experience
  • No new equipment – compatible with all IFE systems
3
AVX Inflight Sound  Overview 2
  • Psycho-acoustic adaptive encoding
  • Parses, disassembles, restructures the dynamic range in real time – constantly adapts to program content
  • Enables full optimization of dynamic range to needs of human hearing without severe limitations of existing technology


4
AVX Inflight Sound  Overview 3
  • Adaptive Encoder vs. Compressor/Limiter
  • DSP intensive vs. analog design
  • Unlimited design possibilities vs. practical limitations
  • AVX “reverse-compression” emulates how brain deals with high ambient noise like conversation on a jet
5
IFE sound system –
The Titanic
  • Destined to cruise over sea of noise – 75db SPL
    (From take-off to landing, like sitting next to a vacuum cleaner, a hair dryer, or standing in the shower.)
  • Can not change course very quickly – prohibitive costs
  • Needs to accommodate all kinds of people and stuff – variety of headphones, hardware, media
6
Film sound track –
The Iceberg
  • Only the peaks are noticed


  • Most of it obscured by sea of noise
  • “Big picture” missing – no context
  • Tells how to perceive visual (picture)
  • Controls emotional response eg. fear
  • Conveying the emotional content is life or death matter to a film
  • Low emotion = low immersion


7
How much is hidden?
  • Top of iceberg is 90db
  • 15db is revealed


  • Sea of noise level 75db


  • 35db  obscured by sea of noise
  • Over 2/3 of sound
    (dynamic range) is buried beneath sea of noise
  • Bottom of iceberg is >40db
8
Human hearing 1


  • Normal range 35-72db for speech (intelligibility)
  • Below 75db  sea of noise in cabin
  • = 0% intelligibility
  • Good news - No need to have foreign language tracks
9
Human hearing 2


  • Lose intelligibility if average speech less than 15 db above noise – 75 + 15 = 90db
  • Narrow range for speech 15 db – average at 90db (unsafe), peaks at 97db
  • Bad news - Above 85db – no increase in intelligibility – instead oppression and discomfort
10
How to uncover the iceberg?
  • Can not change noise level
     - 75db
  • Iceberg is 50db high  (dynamic range) 90db-40db
  • Can not lift above noise level 5db+75db=125db !!!!!
  • 90db beginning of “unsafe levels” (D.O.L.), discomfort
  • 2 hours @ 100db = hearing damage – long flights?
11
Existing technology ineffective 1
  • Squeezes to 30-40db range
    • Ear discomfort from peaks
    • Obscured mood fx / underscore
    • Straining to understand – low intelligibility
  • Jet noise “in front” of sound track
    • Like looking through dirty glass
  • Constant volume adjustments
  • Little or no immersion in film
12
Existing technology ineffective 2
  • Operates from top down
    • Characteristics of electronics
    • Compressor
    • Limiter
  • Unnatural
    • Real life is opposite
  • Limited by artifact -claustrophobic
13
Human hearing psychology 1
  • Decoding, decoding, decoding
    • Sound into phonemes
    • Phonemes into words
    • Words into language
    • Language into meaning
  • Brain overwhelmed
    • Sound buried in noise
    • Playing “catch-up” -Trying to figure out what was missed while trying to continue in-the-moment decoding
14
Human hearing psychology 2

  • Failure & frustration
    • No one likes to feel stupid
    • Not knowing “what’s going on”
    • Missing vital information that “others know”
  • Ability to make emotional associations is blocked by obsession to decode
15
Brainy brain has no problem
  • Why we can have a normal conversation above the noise?
  • Brain uses “reverse-compression” to raise bottom with little affect on top
  • High intelligibility with acceptable  levels
  • Normal speech pattern -subconsciously adapt to
    ambient noise
16
AVX one-ups
the brain 1
  • Extreme
    reverse-compress/limit
    micro-encodes detail
  • Top is normal – ‘must be normal’
  • Comfortable levels
  • Dynamic range fits
17
AVX one-ups
 the brain 2

  • Raises bottom of sound track above noise
  • Mask effect – sound track is relentlessly dominant
  • Mind tunes out and soon forgets subordinate jet noise (it wants to)
18
AVX one-ups
 the brain 3

  • Enhanced aural cues ~100% intelligibility including spatial cues
  • Sound decoding easy & relaxing
  • Brain resources free to make deep emotional associations
  • High level of immersion in film
  • Tune out flying – trip is shorter and more enjoyable
19
Film Clips from
SECONDHAND LIONS comparing
WAEA 1289-1 spec. audio with
AVX encoded audio
  • Clips mixed with jet noise @~73db for simulation
  • All clips are at the same volume level
  • Louder sounds are similar
  • AVX raises quiet sounds


20
Per 1289-1 spec. #1
  • Inaudible under bridge, cop, plane approach
21
AVX Encoded #1
  • Distinct snoring & ambience under bridge, plane approach
22
Per 1289-1 spec. #2
  • Inaudible off camera speech, music and ambience
23
AVX Encoded #2
  • Speech is intelligible,
     distinct music and ambience
24
Per 1289-1 spec. #3
  • Inaudible lion growls,
    weak ambience,
    lack of presence
25
AVX Encoded #3
  • Distinct lion growls,
    ambience is dominant, speech has good presence


26
Per 1289-1 spec. #4
  • Inaudible bell-tree, no presence to music and ambience
27
AVX Encoded #4
  • Distinct bell-tree, good presence of music and ambience
28
Per 1289-1 spec. #5
  • Inaudible ambience, unclear speech at end
29
AVX Encoded #5
  • Better ambience, distinct speech at end “learn to do without” & “uh-huh”
30
Per 1289-1 spec. #6
  • Speech is frustrating - unclear, music missing until end
31
AVX Encoded #6
  • Speech is intelligible,
     music is marginally audible
32
Per 1289-1 spec. #7
  • Speech is marginal, jet noise dominates
33
AVX Encoded #7
  • Speech is crisp and present, music and ambience dominate over jet noise
34
Per 1289-1 spec. #8
  • Jet noise dominates,
    missing/obscure lion sound,
    ambience very weak
35
AVX Encoded #8
  • Sound track dominates,
    lion & ambience clear,
  • presence, “in the movie”
36
Per 1289-1 spec. #9
  • Jet noise dominates,
    speech lacks presence,
    ambience very weak
37
AVX Encoded #9
  •      Track dominates, speech is intimate & clear, ambience, reverb, presence, “in the movie”
38
Summary

  • AVX solves the difficult noise problem for IFE films
  • Film perceived as intended
  • High immersion > enjoyment > successful distraction
  • Greater satisfaction & positive impression of flight
39
AVX 757 Simulator Demo

  • 5-8 after show, 10 minute drive west near Disney & Warner Bros.
  • Sound Trax Studios
    2815 West Burbank Blvd.
    Burbank 818-842-6300
40
AVX would like to thank

  • WAEA attendees
  • John Salzman
  • John Eargle & Chris Foreman –sound reinforcement research
  • Avid Airline Products – headphones
  • Sound Trax Studios
  • New Line Cinema