The Basis for the transcription of speech sounds - PowerPoint PPT Presentation

1 / 45
About This Presentation
Title:

The Basis for the transcription of speech sounds

Description:

kip keep there is usually no puff of air (aspiration) with the 'p' ... divine divinity [aI] [I] profane profanity [eI] [Q] serene serenity [i] [E] 41 ... – PowerPoint PPT presentation

Number of Views:166
Avg rating:3.0/5.0
Slides: 46
Provided by: chssMon
Category:

less

Transcript and Presenter's Notes

Title: The Basis for the transcription of speech sounds


1
Chapter 2
  • The Basis for the transcription of speech sounds
  • Different Types of Transcription
  • Transcription Practice
  • Distinctive Features

2
Some Sounds Differ according to their Environment
  • kip keep there is usually no puff of air
    (aspiration) with the p
  • pik peek there is always aspiration with the
    p
  • Q Is the difference between these two ps
    significant?

3
What are the basic or significant sounds of
English?
  • The significant sounds differentiate meaning
  • 1. aI s? D? mAb
  • 2. aI s? D? mApH (p with aspiration)
  • 3. aI s? D? mAp (p without aspiration)
  • (1) and (2) differ in meaning
  • (2) and (3) do not.
  • (1) and (2) are contrastive.
  • So, p and b represent different phonemes in
    English
  • But p and pH do not.

4
The Phoneme
  • A phoneme is a speech sound that is capable of
    differentiating meaning.
  • The sequence
  • pill bill till dill kill gill
  • indicates that p,b,t,d,k,g are phonemes in
    English

5
Variation within a Phonemeworksheet 1-4
  • Question What about p and pH (aspirated
    p)?
  • Do they represent different phonemes in English?
  • ( Problem 4 shows that p and pH do represent
    different phonemes in Hindi)
  • Answer They do not.
  • p and pH are allophones of the phoneme p
  • p and pH are in complementary distribution
  • where pH occurs, p does not
  • where p occurs, pH does not
  • (Think Superman Clark Kent)

6
Complementary Distribution
  • The distribution of allophones in their
    respective phonetic environments such that one
    never appears in the same phonetic context as the
    other.
  • For example, the p and pH allophones of /p/
    are in complementary distribution in the word
    initial vs. post-s environments pHIn
    and spIn
  • pIn and spHIn

7
Phonemic Notation
  • /p/ phoneme
  • pH p p allophones
  • for p lips remain closed

8
How do we determine the significant sounds of a
language?
  • 1. aI s? D? mAb
  • 2. aI s? D? mAp
  • Significance is determined by contrast, as in
  • mAb versus mAp
  • mAb and mAp are contrastive or minimal pairs

9
The significant sounds are found by testing
minimal pairs
  • pHIl
  • bIl dIl tHIl kHIl
  • gIl fIl vIl 1 mIl
  • wIl Til2 Dil3 sIl
  • zIl4 nIl lIl ?Il
  • Sil tSIl Nil5 dzIl
  • Zil 6 hIl
  • pHil contrasts with bil, and pHil contrasts with
    dil, etc.
  • 1pHaIl vaIl
  • 2 pHIn Tin
  • 3pHaI DaI
  • 4pHA? zA?
  • 5?Ip ?iN
  • 6No contrast for Z vs. p
  • but Z and p are not phonetically similar so
    we assume they are separate phonemes.

10
The Need for an Alphabet
  • We need to represent the contrasting sounds of
    any given language unambiguously
  • We need an alphabet to do this

11
Transcription Which Alphabet?
  • We need unambiguous representation of sound
  • How about the English spelling system?
  • Same sound, different symbols to,too,two
  • Different sounds, one symbol dad, father, call,
    sofa
  • One sound, sequence of symbols tough, physics
  • One symbol, sequence of sounds exit EgzIt
  • Symbols, but no sound pneumonia

12
Phonetic Transcription
  • Conclusion We need a phonetic alphabet
  • with ONE SYMBOL for ONE SOUND
  • Two systems used by linguists
  • American
  • IPA

13
American Transcriptions, z, c, j, y
  • arose from need to transcribe indigenous
    languages of North America
  • a practical system for publication
  • (typographically easy)
  • aims at phonemic transcription
  • with phonetic detail consigned to discussion
    notes
  • intended for American languages only

14
International Phonetic Association/Alphabet
  • founded in France in 1886
  • an organization for teachers of language
  • originally
  • aims at a system that will represent all
    languages

15
The International Phonetic AlphabetPrinciples of
the IPA
  • a separate letter for each distinctive sound
  • universal use of one symbol for the same sound
    across languages
  • use of ordinary letters of roman alphabet where
    possible
  • alphabet should accord with phonemic principle
    and cardinal vowel system
  • diacritic marks only for
  • suprasegmentals
  • non-meaningful distinctions
  • minute shades of sound for scientific purposes

16
English Articulation of b(Lab 3)
lips
Silence (no v. cord vibration)
b

v.cord vibration
LABM030
17
English Articulation of ph(Lab 3)
lips
Silence (no v. cord vibration)
p

v.cord vibration
LABP030
h (aspiration)
18
Hindi Articulation of p(Lab 3)
lips
Silence (no v. cord vibration)
p

v.cord vibration
(no aspiration)
19
IPA Non-roman Consonant Symbols
  • N eng
  • T theta
  • D eth
  • S esh
  • Z yogh
  • ? turned r
  • Pullum Ladusaw. Phonetic Symbol Guide. U of
    Chicago Press.

20
IPA non-Roman Vowel Symbols
  • E epsilon
  • Q ash
  • U upsilon
  • ? open o
  • A script a
  • ? turned v
  • ? schwa

21
Notational Conventions in Transcription
  • / / encloses phonemic transcriptions
  • encloses phonetic (allophonic)
    transcriptions
  • denotes a word boundary
  • denotes a morpheme boundary
  • V stands for any vowel
  • C stands for any consonant
  • C0 stands for a sequence of zero or more
    consonants

22
Transcription Types
  • Phonemic Transcription
  • ? bEt? pA?t
  • Broad Phonetic Transcription
  • ? bER? pHA?t
  • Narrow Phonetic Transcription
  • ? bER? pHA?t
  • WWW

23
Relations between Speech Soundsworksheet Free
Var.
  • Relation Element Example
  • contrast phoneme /p/ vs. /b/
  • complementary allophone p vs. pH pin vs.
    spin
  • distribution
  • free variation phoneme/allophone
  • p vs. pH stop vs. stoph
  • E vs. i
  • economics

24
New Symbols to Capture Variation
  • Symbol Name Articulation
  • ? fishhook r flap
  • ? glottal stop glottal stop
  • ? tilde l velarized l
  • l6 under-ring devoicing
  • n syllabicity mark same as n

25
Doing Phonemic Analysis
  • Determining the relationship of 2 sounds
  • Are there minimal pairs pQt vs. bQt
  • if yes, then the relation is contrast
  • If no, do the sounds differ predictably
  • pHIn vs. spIn
  • if yes, then the relation is complementary
    distribution. Determine the basic allophone.
  • If no, the sounds must be in free variation.
  • Phonemic Analysis Problems

26
The Trick of Phonemic AnalysisList all
environments of the sound in question
  • Burmese
  • Voiced nasals occur Voiceless nasals occur
  • ___ i h ___ i
  • ___ w h ___ w
  • e ___ h ___ y
  • ___e h ___ e
  • ___ w h ___ w
  • The voiceless nasal occurs after . . .
  • The voiced nasal occurs . . .

27
The Elsewhere Condition
  • a principle governing two overlapping rules which
    dictates that the more specific rule should be
    tried first followed by the more general rule
  • EXAMPLE In Burmese,
  • - nasals are voiceless after /h/,
  • - elsewhere they are voiced.

28
The Notion of a Natural Class
  • Speech sounds can be described by articulatory
    features
  • t - a voiceless alveolar stop
  • m - a voiced bilabial nasal
  • Sounds can be grouped by features
  • Voiced stops b,d,g
  • Labials p,f,b,v,m,w
  • We call such a grouping a natural class of sounds

29
Natural Classes of Sounds
  • Capture generalizations about
  • the sound systems of a language
  • e.g. English plural s vs. z relies on
    voicing
  • dialect variation
  • the pronunciation of borrowed words
  • first language acquisition
  • processes of sound change
  • worksheet Natural Classes

30
Background on Distinctive Features
  • Until the 1960s, the phoneme prevailed as the
    basic unit of the analysis of sound
  • The system that treated the phoneme as the basic
    unit was called Taxonomic Phonemics
  • in the 60s, linguists (in particular Noam
    Chomsky and Morris Halle) pointed out
    difficulties with the rigid system of T.P.
  • we will now look at one of these difficulties

31
Taxonomic Phonemics
  • developed by American linguistics
  • divorced sound from meaning
  • assumed certain principles that would lend a
    scientific precision to phonemics

32
The Phoneme
  • A phoneme is a family of similar sounds which
    language treats as being the same.
  • If there is a contrast between two sounds in one
    environment, then these two sounds must be
    considered different phonemes in all
    environments.
  • Bernard Bloch. Phonemic overlap.

33
In most environments, the presence of A and
A? is predictable
  • The long/short distribution in
  • pAt pA?d
  • lAk lA?g
  • mAp mA?b
  • indicates that A and A?are allophones
  • in complementary distribution

34
But in some words, A vs. A? is the only
contrasting sound
  • The contrast of /A/ and /A?/ in
  • bomb bAm balm bA?m
  • indicates that /A/ and /A?/ are phonemes in
    contrast
  • Contrast also in
  • sorry starry
  • bother father

35
Asymmetry in Phonemic Representations
  • To account for the phonemic contrast of
  • bomb bAm balm bA?m
  • Taxonomists had to claim that the distinction in
    pAt pA?d
  • was also phonemic even though the variation is
    predictable from context

36
Features
  • Chomsky pointed out that we could have both the
    predictability of pAt pA?d and the contrast
    of bAm bA?m by describing the sound variation
    with
  • a unit smaller than the phoneme (the feature)
  • a lexicon that contained underlying features
  • rules to describe the variation
  • Chomsky, N. 1964.
  • Chomsky, N. M. Halle. 1967.

37
A Feature Account of the A A? Distinction
  • Lexicon Phonological Rule Pronunciation
  • /bAm/ bAm
  • /bA?m/ bA?m
  • /pA?t/ V?-long/__-voiced pAt
  • /pA?d/ pA?d
  • -long, -voiced are called distinctive features

38
A Feature Matrix in the Lexicon
  • A word is represented as a Feature Matrix

39
Distinctive Features
  • Many features are based on articulation
  • labial coronal dorsal
  • Some are based on acoustics
  • strident high sonorant
  • Most are based on a combination of articulation
    and acoustics
  • round lateral continuant

40
Advantages of Distinctive Features
  • Features allow the statement of generalizations
  • V ? V / _________ C
    V
  • tense ?tense -stress
  • divine divinity aI ? I
  • profane profanity eI ? Q
  • serene serenity i ? E

41
Advantages of Features
  • Features represent allophonic variation as a
    systematic rather than a random process
  • English V ? V / _____ -Voiced
  • -long
  • Features directly reflect the articulatory and
    acoustic activity involved in each sound

42
Features as Universals
  • Currently about 24 are used
  • they constitute a claim about what is possible in
    human phonological behavior
  • the binary (/-) nature of the features plays
    into claims of universality

43
Feature OppositionsRoman Jakobson, 1941
  • Silence
  • -consonantal consonantal
  • high -high oral -oral
  • -back back labial -labial labial -labial
  • apply in language acquisition
  • apply in language dissolution
  • correspond to the frequency of sounds in the
    worlds languages _

44
Feature Oppositions
  • Stage 1 /-consonant papa or baba
  • Stage 2 /-oral papa, mama
  • Stage 3 /-labial mama/nana, papa/tata
  • Stage 4 /-high papa, pipi
  • Stage 5 /-back pipi, pupu

45
References
  • Chomsky, N. 1964. Current Issues in Linguistic
    Theory.
  • Chomsky, N. M. Halle. 1967. The Sound Pattern
    of English.
  • Jakobson, Roman. 1941. Childs Speech, Aphasia,
    and Linguistic Universals.
Write a Comment
User Comments (0)
About PowerShow.com