Speech Production & Perception

• Early stages of speech development (Ryalls & Behrens)

  Speech language developmental Stages: Ryalls & Behrens

  1. Prelingual stage (0-6 months)
  2. Babbling stage (6-12 months)
  3. One - Word or Holophrastic Stage (1-2 yrs)
  4. Two - Word Stage (~2 yrs)
  5. Telegraphic stage (not age; MLU)
• Prelingual stage (0-6 months)

  • Nonlinguistic 
  • Non language sounds
  • Eye contact
  • Turn taking evident 
  • Developing native prosody
• Babbling stage (6-12 months)

  • Practice articulation
  • Exploration of various articulators and coordination among them
  • Canonical babbling /ba-ba-ba/
  • Variegated babbling /ti-ta-ku/
  • Conversational babbling 
  • Protowords: phonetically consistent forms
• One-Word or Holophrastic Stage (1-2 yrs)

  • Mostly CV/VV
  • Vocabulary development
  • Naming explosion or spurt around 18 months
• Two-Word Stage (~2 yrs)

  • Concept of agent-object-action emerging 
  • Use only combination of two roles
  • Agent-object [mama book]
  • Agent-action [mama sing]
  • Action-object [doggie run]
• Telegraphic stage (not age; MLU)

  • Lengthen two-word utterances by using conjunctions, modifiers, location etc
  • Morphological development 
  • Phonological processes are slowly eliminated
  • The dog/s are run/ing = 6
  • Doggie run  = 2
• Prelinguistic Vocalizations

  Reflexive crying and Vegetative Sounds (0-6 weeks)

  Cooing & Laughter (6-16 weeks)

  Vocal Play (16-30 weeks)

  Canonical Stage (31-50 weeks)
• Prelinguistic Vocalizations

  Reflexive crying and Vegetative Sounds (0-6 weeks)

  Grunts 

  • Grunts are defined as relatively short, monosyllabic utterances.
  • They are characterized by an abrupt glottal onset followed either by a syllabic nasal, or most often nasalized vowels.
  • A grunt is a vegetative - sounds like burping and coughing 
  • Communicative grunts however are produced intentionally by the child during the interaction with an adult.
• Prelinguistic Vocalizations

  Cooing & Laughter (6-16 weeks)

  Laughter 

  • Laughter is predominantly conceptualized as vowel-like bursts
  • The vowels are typically close to center of their vowel space.
  • It is rich in harmonics
  • Structure of laughter is highly consistent within a series
• Prelinguistic Vocalizations

  Vocal Play (16-30 weeks)

  Vocal Play

  • Known as expansion stage or exploratory mapping stage by Oller.
  • It is characterized by playful use of behaviors such as squealing, growling, yelling, whispering, and production of noises by blowing air or food or saliva through constrictions in mouth or pharynx (raspberries).
• By 4 months:

  • Larynx begins to drop
  • Presence of pharyngeal cavity
  • Lowering and raising the jaw

  Exploratory Stage 

  • There is increasing control of laryngeal and oral articulatory mechanisms
  • An infant can make varied contractions with PHARYNX 
  • Jaw lowering allows for varied tongue contractions and tongue lowering.
  • Jaw raising allows for varied lips configurations
  • At this stage infants have increased to manipulate prosodic aspects of speech.
  • Some vocal behaviors at this stage include: raspberries, squalls, growls, yells, whispers
• Prelinguistic Vocalizations

  Canonical Stage (31-50 weeks) 

  • Canonical stage is characterized by well-timed, closing and opening sequences with normal phonation.
  • Generally, after about 6 mos vocal behaviors are perceived to be adult-like CV syllables, oscillating between articulatory closure and opening (babbling).
• Vowel Perception/Production

  • Produced when the buzzing sound from vocal fold vibration is modified by the resonances of the vocal tract.
  • Different vowels have different vocal tract configurations and therefore different resonances
  • Formant frequencies are important in perception of vowels, especially the first three formants (F1, F2, and F3). It is possible however to perceive vowels when the first two formants are missing.
• Various cues that aid in the perception of vowels, stops, fricatives and nasal consonants 

  Vowel Perception/Production

  • Range of formant frequencies for a particular vowel generally overlaps with those of other vowels.
  • There is no acoustic feature that completely identifies a speech sound. Knowledge of a speaker, linguistic context, our expectations, and rules of grammar all influence how we perceive speech.
• Various cues that aid in the perception of vowels, stops, fricatives and nasal consonants

  Perception/Production of Fricative Consonants 

  • Fricatives are produced by air turbulence created when air from lungs is forced through a vocal tract constriction formed by the lips or tongue.
  • Fricatives are distinguished from each other based on:
  • Intensity of frication noise
  • Spectral differences (frequency range)
  • Formant transitions of their neighboring vowels
  • Duration of the frication (e.g. stops versus fricatives)
• Various cues that aid in the perception of vowels, stops, fricatives and nasal consonants

  Perception/Production of Nasal Consonants 

  • Nasal consonants are produced when the soft palate (velum) is lowered to allow air to pass through the nasal cavity. The air flow through the oral cavity is completely stopped using the lips or tongue.
  • The amplitude / intensities of the formants lowered
  • Weak nasal resonances are present close to first formant frequencies
  • The nasal consonants ([m], [n] and [ng]) are distinguished based on the second formant transitions of the following vowel
• Various cues that aid in the perception of vowels, stops, fricatives and nasal consonants

  Perception of Stop/Plosives/Burst 

  • A plosive or burst is produced by closing the vocal tract using our lips or tongue to stop the flow of air completely and then suddenly releasing the built-up air pressure.
  • The plosive consonants in American English are p, t, k, b, d, & g. Also called as stop consonants
  • The perception of any plosive consonant depends on the frequency of the plosive and the formant frequency of the following vowel.
• Various cues that aid in the perception of vowels, stops, fricatives and nasal consonants

  Perception of Stop/Plosives/Burst 

  • Plosive consonants can also be heard in the absence of plosive burst, when the frequency of the second formant has been tilted up or down, during the initial segment of the syllable.
  • This portion is called the second formant transition.
  • There is no single target or cue (e.g., second formant frequency) for the perception of any one plosive consonant.
• Bottom - up Perception

  Importance to signal coming in air

  bottom up refers to theorists and researchers telling us that speech is dependant that we are hearing; because all the information that we need is from the signal

  
  

  Gives importance to spoken language that enters the ear and how it is processed phonetically
• Top-down Perception

  Our knowledge about the speech sounds, which we apply when we hear something

  top down - we still use linguistic knowledge to make decisions

  Knowledge about the universe and linguistic context is used during top-down perception
• McGurk effect

  • Harry McGurk, senior developmental psychologist, university of Surrey in England.
  • Audio syllable “ba” dubbed onto a visual “ga” is perceived as “da”
  • This audio-visual illusion has become known as the McGurk effect.
  • This illusion demonstrates interaction between articulatory and auditory information. Highlights importance of visual information in hearing.
  • States that vision is equally important for auditory perception
• Motor theory of speech perception

  • This theory proposes that listeners continuously articulate the incoming speech sounds and compare the auditory results of their own articulation with incoming auditory patterns.
• What is the Motor Theory of Speech Perception?

  • This theory hypothesizes that people perceive spoken words by identifying the vocal tract gestures with which they are produced rather than by identifying the sound patterns that speech generates.
  • The role of the speech motor system is not only to produce speech but also to detect them.
• The three main claims of the theory are the following:

  1. Speech processing is special:  Depends on the DISTAL GESTURES that generate the acoustic patterns; NOT the proximal acoustic patterns. It amplifies recruitment of the motor system, in perception.
  2. Perceiving speech is perceiving vocal tract gestures: Vocal tract gestures are the objects of speech perception. The McGurk effect goes along with this claims and is the result of the prototypes in memory that specify the optical and acoustic cues for syllables and is involved in perceptual tasks.
• The three main claims of the theory are the following:

  1. Speech perception involves access to the speech motor system - Holds that we recognize the sounds of speech by creating a motor representation of how those sounds of speech by creating a motor representation of how those would be produced.
• Evidence against the Motor theory of speech perception.

  • People with Broca’s Aphasia have problems with speech motor control and therefore unable to produce intelligible speech but understand speech nevertheless.
  • People with Wernicke’s Aphasia have difficulty understanding spoken language but can execute speech motor commands.
• Evidence for the Motor theory of speech perception.

  • People who speak foreign languages often mimic lip movements to better comprehend other languages.
  • The McGurk Effect - listeners combine what they see with what they hear and end up hearing something in between.