English Phonetics and Phonology
The Speech Chain
Lecture 3 The Speech Chain
Peter B. Denes & Elliot N. Pinson (1963/1971)
The Speech Chain: The Physics and Biology of
Spoken Language
KAMIYAMA, Takeki
[email protected]
The Speech Chain
• What are the necessary conditions for
speech communication to take place?
(acoustic)
2015-2016 S1 T. Kamiyama
Phonetics and Phonology Lecture 3
3
Speaker
2015-2016 S1 T. Kamiyama
signal
Phonetics and Phonology Lecture 3
Listener
4
The Speech Chain
Then, three conditions …
Speaker
Listener
(acoustic)
• Savoir (+ savoir-faire)
• Pouvoir
• Vouloir
signal
Encoding
2015-2016 S1 T. Kamiyama
Phonetics and Phonology Lecture 3
5
Decoding
From Denes and Pinson The Speech Chain
2015-2016 S1 T. Kamiyama
Phonetics and Phonology Lecture 3
6
Stage 1: Conceptualization
Stage 2: Linguistic encoding
Speaker
Speaker
Input: psychic impulse
This encoding device needs linguistic
knowledge:
Savoir
Semantics and pragmatics
Vocabulary
Morphology
Pouvoir
Syntax
Phonology (vowels and consonants,
prosody)
Vouloir
Encoding
Output: preverbal
message
Encoding
From Denes and Pinson The Speech Chain
2015-2016 S1 T. Kamiyama
7
Phonetics and Phonology Lecture 3
Input: preverbal message
Output: phonetic sequence
(including prosody)
From Denes and Pinson The Speech Chain
2015-2016 S1 T. Kamiyama
8
Phonetics and Phonology Lecture 3
Stage 3: Physiological encoding
Stage 4: Acoustic transmission
Speaker
Speaker
Input: phonetic sequence
(including prosody)
Listener
(acoustic)
signal
Pouvoir
(acoustic)
signal
Encoding
2015-2016 S1 T. Kamiyama
The brain orders the
appropriate muscles of the
appropriate speech organs
to move properly, through
motor nerves. Savoir-faire
The result: the speech
organs produce sound.
Output: sound (acoustic
signal)
9
Stage 5: Physiological decoding
Output: auditory cues
(acoustic)
Listener
signal
The human ear decodes the acoustic
signal into different frequency zones.
The information is sent to the brain
through sensory nerves.
Input: sound (acoustic
signal)
2015-2016 S1 T. Kamiyama
Encoding
From Denes and Pinson The Speech Chain
Phonetics and Phonology Lecture 3
Pouvoir
Decoding
From Denes and Pinson The Speech Chain
2015-2016 S1 T. Kamiyama
10
Phonetics and Phonology Lecture 3
Stage 6: Linguistic decoding
Output: preverbal
message
This decoding device needs linguistic
knowledge:
Phonology (vowels and consonants,
prosody)
Savoir-faire
Syntax
Vocabulary
Savoir
Morphology
Pouvoir
Semantics and pragmatics
Input: auditory cues
Decoding
From Denes and Pinson The Speech Chain
11
Phonetics and Phonology Lecture 3
Movement of particles is
transmitted through a
medium such as the air,
water, etc., and reaches the
listener’s ear.
Listener
Decoding
From Denes and Pinson The Speech Chain
2015-2016 S1 T. Kamiyama
Phonetics and Phonology Lecture 3
12
Stage 7: Comprehension
Output: psychic pulse
Feedback circle
Listener
Speaker Decoding
Feeling that the listener has understood
something.
Input: preverbal message
(acoustic)
signal
Decoding
Encoding
From Denes and Pinson The Speech Chain
2015-2016 S1 T. Kamiyama
Phonetics and Phonology Lecture 3
13
The speaker listens to
his/her own speech,
understands it at the
same time; monitors
his/her production, then
adjusts it if necessary.
From Denes and Pinson The Speech Chain
2015-2016 S1 T. Kamiyama
Phonetics and Phonology Lecture 3
14
Branches of phonetics
Speaker
Articulatory phonetics studies how
speech sounds are produced by the
organs of speech.
2. Phases of the Speech Chain
and branches of phonetics
1. Traditional approach: observation,
reproduction and introspection.
2. Instrumental and experimental
approach: use of instruments such as
camera, video, endoscope,
electrodes, sensors of air pressure,
brain imaging, etc.
Encoding
2015-2016 S1 T. Kamiyama
From Denes and Pinson The Speech Chain
Phonetics and Phonology Lecture 3
16
Ultrasound imaging (ecography)
Queen Margaret University, Edinburgh
http://www.qmu.ac.uk/casl/ultra/subjectinfo_ESRC_2007_2008.htm
2015-2016 S1 T. Kamiyama
Instruments for articulatory
investigation (Laboratory of
17
Phonetics and Phonology
Lecture 3
phonetics
and phonology, Paris)
2015-2016 S1 T. Kamiyama
Phonetics and Phonology Lecture 3
18
New development?: ePGG
What is PGG?
Sawashima & Hirose (1981), In Vocal Fold Physiology
Honda & Maeda
Transillumination
technique
photodiode
External lighting and sensing PGG (ePGG)
The ePGG sensor detects transillumination from
a high-power LED placed on the neck skin.
PGG signal
k
2015-2016 S1 T. Kamiyama
e:
Phonetics and Phonology Lecture 3
k
e:
19
2015-2016 S1 T. Kamiyama
Non-invasive method
Branches of phonetics
Very experimental
ePGG + Airflow + Pio setup
(acoustic)
signal
2015-2016 S1 T. Kamiyama
Phonetics and Phonology Lecture 3
20
Phonetics and Phonology Lecture 3
21
Acoustic phonetics
studies the acoustic,
physic characteristics
of speech sounds, by
using instruments
such as oscillograph
or spectrograph.
From Denes and Pinson The Speech Chain
2015-2016 S1 T. Kamiyama
22
Phonetics and Phonology Lecture 3
Branches of phonetics
Sound spectrogram
(or sonagram, sonogram)
Auditory phonetics studies how
listeners decode and understand
speech sounds.
1. Psycholinguistic approach:
analysis of the behaviour (task of
choice, judgement, etc.) in response
to speech sounds, or synthesised
sounds.
Listener
Decoding
2. Neuroscience approach: direct
observation of the brain by brain
images obtained by MRI, EEG, etc.
2015-2016 S1 T. Kamiyama
Phonetics and Phonology Lecture 3
23
From Denes and Pinson The Speech Chain
2015-2016 S1 T. Kamiyama
Phonetics and Phonology Lecture 3
24
Perception test
• Identification test
• Discrimination test
End of Lecture 3
Lecture 4: Organs of speech
2015-2016 S1 T. Kamiyama
Phonetics and Phonology Lecture 3
25