Language and Mind, Week 4
Language and Mind, Week 4 LING 275
Popular in Language and Mind
Popular in Linguistics
This 2 page Class Notes was uploaded by Carole Boulware on Wednesday July 20, 2016. The Class Notes belongs to LING 275 at University of Southern California taught by Elsi Miia Kaiser, Rachel Walker in Spring 2016. Since its upload, it has received 12 views. For similar materials see Language and Mind in Linguistics at University of Southern California.
Reviews for Language and Mind, Week 4
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 07/20/16
LING 275 Language and Mind, Spring 2016, Kaiser/Walker Speech perception: Lack of Invariance [Overview of March 10, 2016] Decoding the Speech Signal: What was the intended message? What were the intended words? What were the speech sounds? Challenges of speech perception • #1 Speaker variability • #2 Segment variability (Lack of invariance: “The acoustic signal for a given segment of speech varies depending on the context in which it is produced”, see course reader) • #3 Segmenting the speech stream into words [after spring break] PLAN FOR TODAY: Challenges of speech perception due to variance: (i) Speaker variability (ii) Segment variability (iii) What remains constant across all this variation? Importance of perception, interpreting the sound wave: Connection between acoustic input and what we perceive is not always straightforward Humans are very good at speech recognition! (Comparison with computers) How do we go from the sound wave to word recognition? One idea: look at formants and try to figure out what phonemes they map on to. BUT this is not going to work: Variability: (i) Speaker variability, (ii) Segment variability (‘lack of invariance’) Speaker variability: Variability between speakers • Fundamental Frequency • Formant Frequencies • Speaking Rate • Accents Variability within a speaker [also across speakers] • Emotional state • Speaking rate • Position of word within sentence Segment variability • Stress (emphasis): changes duration, amplitude, pitch (e.g. an INsult, to inSULT) • Coarticulation (see last week’s lectures, cape cod example) So, what is stable across all this variation? What are the cues in the acoustic signal that tell you what phoneme you’re hearing? o Vowels Formats are crucial for determining vowel quality Importance of the relative values of F1, F2, F3 (see also earlier lectures) o Stop consonants Importance of formant transitions, formant transitions vary according to place of articulation (see also earlier lectures) Example with /di/ and /du/, direction of the ‘tail’ depends on formants of following vowel There is no one-to-one relationship between F1&F2 formant/transition pattern and [d] BUT the locus of F2 remains constant (F2 ‘points to’ about 1700-1800 Hz) Importance of the transition between the consonant and vowel articulation. In the left figure, only the second formant transition is changing. What do you hear? (b/d/g) [See also Figure 5.8 on page 134 of the reader] The Lack of Invariance Problem: The acoustic signal for a given speech segment varies depending on the linguistic context in which it is produced. There is no one-to-one relationship between the sound wave and the phoneme you perceive. Phoneme perception is a guessing game: What’s the most likely phoneme given a bunch of semi-informative cues? Importance of interpreting the sound wave: Sound wave => ear => mind/brain
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'