The Cocktail Party Effect

1
The Cocktail Party Effect

• Presented by Group 8
• ??? 93121305
• ??? 93121307
• ??? 93121311
• ??? 93121327
• ??? 93121335
• ??? 93121337
• ????? 93121375
• ??? 91117054

2
The Cocktail Party Effect

• The Inventor
• Definition
• Cause
• Auditory System
• Recognition of Speech With One and Two Ears
• Source Separation
• Automatically Recognizing Streams
• Conclusion and Games

3
The Inventor of The Cocktail Effect

• Colin Cherry
• 1953
• Air traffic controllers
• 1950's

4
About The Experiment

• Separate two sounds from a single loudspeaker at
  the same time
• gender, direction, speed, pitch

5
Broadbents Filter Theory

• Our mind can be conceived as a radio receiving
  many channels at once
• Which channel we should pay attention ?

6
The Cocktail Party Effect

• The ability to focus one's listening attention on
  a single talker among a mixture of conversations
  and background noises, ignoring other
  conversations.

7
When Does The Cocktail Party Phenomenon Occur?

• 1.When we are paying attention to one of the
  sounds around us.
• 2.When it is invoked by a stimulus which grabs
  our attention suddenly.
• three times louder than the ambient noise.

8
Cause
Cause

• There is a filter in our brain which can select
  which message we should pay attention to.
• Ex. If someone call your name in the other side
  of the party room calls out your name, we can
  notice that sound and response it immediately.

9
Auditory System The Sensory System For The Sense
of Hearing

• Outer ear, Middle ear, Inner ear
• Two ears can identified the location of the sound
  limit 20 Hz to 15kHz.distinquish less than 1.5
  Hz
• The auditory system filters sounds received,
  ignores background noise and concentrate on
  important information
• Ex. mothers to the cries of her baby.
• The ear can differentiate quite subtle sound
  changes and familiar sounds.

10
Recognition of Speech With One and Two Ears

• Different directions
• Lip-reading, gestures, and the like
• Different speaking voices
• Different accents
• Transition probabilities

11
Recognition of Speech With One and Two Ears

• No difficulty in listening to one ear
• A slight delay
• Temporary memory

12
Source Separation

• Source separation problems in digital signal
  processing are those in which several signals
  have been mixed together and the objective is to
  find out what the original signal were.
• Principal Components Analysis (PCA)
• Independent Components Analysis (ICA)

13
Automatically Recognizing Streams

• It is generally difficult to find tractable and
  accurate computational solutions to recognition
  problems than human find simple.
• Evidence-partition process
• Experiment in 1971