1. The Perception of Speech

2. Speech is for rapid communication Speech is composed of units of sound called phonemes –examples of phonemes: /ba/ in bat, /pa/ in pat Speech

3. Speech can be characterized by a spectrogram Acoustic Properties of Speech

4. Spectrogram reveals differences between phonemes The differences are in the formants and the formant transitions Acoustic Properties of Speech

5. Perceiving Speech So perceiving (interpreting) speech sounds is simply a matter of matching the spectrotemporal properties (the shape of the spectrogram) of the incoming sound waves to the appropriate phoneme right?…

6. Perceiving Speech So perceiving (interpreting) speech sounds is simply a matter of matching the spectrotemporal properties (the shape of the spectrogram) of the incoming sound waves to the appropriate phoneme Then specific phonemes must correspond to specific spectrograms - a property called acoustic-phonetic invariance

7. Acoustic - Phonetic invariance says that phonemes should match one and only one pattern in the spectrogram –This is not the case! For example /d/ followed by different vowels: Perceiving Speech

8. Acoustic - Phonetic invariance says that phonemes should match one and only one pattern in the spectrogram –This is not the case! For example /d/ Clearly perception and understanding of speech sounds is more elaborate than simply interpreting an internal spectrogram Perceiving Speech

9. The phrase “Peter buttered the burnt toast” has five /t/ phonemes. There are not 5 identical sweeps in the spectrogram Perceiving Speech

10. The Segmentation Problem Segmentation is the perception of silence between words Often illusory Perceiving Speech

11. The phrase “I owe you a Yo-Yo” has no silence in it ! Perceiving Speech

12. Spoken Input The Segmentation Problem: –The stream of acoustic input is not physically segmented into discrete phonemes, words, phrases, etc. –Silent gaps don’t always indicate (aren’t perceived as) interruptions in speech

13. Spoken Input The Segmentation Problem: –The stream of acoustic input is not physically segmented into discrete phonemes, words, phrases, etc. –Continuous speech stream is sometimes perceived as having gaps

14. Perceiving Speech So how do you perceive speech? Some of the “strategies”: 1. reduce the data 2. use context clues 3. use vision

15. Categorical Perception is a phenomenon in which the brain assigns a stimulus into one or another category but never into an intermediate category Categorical Perception

16. For example, /ba/ and /pa/ differ in their formant transitions –/ba/ is formed by stopping the flow of air from the lungs and releasing it after about 10 milliseconds (called voice onset time) –/pa/ is similar except that voice onset time is about 50 ms Categorical Perception

17. Voice onset time can range from zero to >50 ms. For example, you could synthesize a sound with a voice onset time of 30 ms but... Categorical Perception

18. Voice onset time can range from zero to >50 ms. For example, you could synthesize a sound with a voice onset time of 30 ms but... English speakers will hear either /ba/ or /pa/ but never something in between Categorical Perception

19. Categorical Perception is Part of Learning a Language Babies can discriminate /ba/ from /pa/ and can discriminate these from phonemes with intermediate voice onset times! By 10 to 12 months, babies (learning English) stop discriminating intermediate voice onset times

20. Categorical Perception is Part of Learning a Language Once category boundaries are learned it is impossible to unlearn them –non-native speakers of any language often cannot hear certain phonemes the way native speakers do –as a consequence they will always have at least some slight accent

21. Another example: Categorical Perception

22. Perception (of all types) Makes Use of Context The stream of information contained in speech is usually ambiguous and incomplete Your brain makes a “best guess” based on the circumstances

23. Perception (of all types) Makes Use of Context Consider the following example: “The __eel fell of the cough shoe”. car”.

24. Perception (of all types) Makes Use of Context Consider the following example: Listeners report hearing the “appropriate” phoneme during the cough “The __eel fell of the cough shoe”. car”.

25. Why rely on only one sensory system when there is information in two !? Much of Speech Perception isn’t Auditory !

26. Why rely on only one sensory system when there is information in two !? The brain seamlessly integrates any information it is given - this is called cross- modal integration Much of Speech Perception isn’t Auditory !

27. Speech perception involves the synthesis of vision and hearing The McGurk effect demonstrates the critical role of vision on speech perception Cross-modal Integration

28. The McGurk Effect Cross-modal Integration

29. The McGurk Effect - suggests that visual and auditory information are combined to enhance speech perception under normal circumstances When visual and auditory information are incongruous the resulting perception is unpredictable and often wrong Cross-modal Integration

30. Auditory Scene Analysis Sounds don’t happen in isolation, they happen in streams of changing frequencies How does the system group related auditory events into streams and keep different streams separate?

31. Auditory Scene Analysis Solving this problem is called Auditory Scene Analysis One important principle is proximity –in pitch, time, or spatial location

32. Auditory Scene Analysis Effect of timing proximity: SlowFast

33. Auditory Scene Analysis Effect of timing proximity: Pitch Do you hear this? Pitch Or this? SlowFast

34. Auditory Scene Analysis Effect of pitch proximity: closefar

35. Auditory Scene Analysis Effect of pitch proximity: Pitch Do you hear this? Pitch Or this? closefar

36. Auditory Scene Analysis Effect of proximity: –auditory system groups together events that happen close together in time and frequency

37. Auditory Scene Analysis Effect of proximity: –auditory system groups together events that happen close together in time and frequency –This enables us to perceive meaningful streams of information when they are mixed with distraction

38. Auditory Scene Analysis Effect of proximity: –auditory system groups together events that happen close together in time and frequency –This enables us to perceive meaningful streams of information when they are mixed with distraction –Interestingly, the brain can disentangle mixed streams only certain circumstances E.g. “The picket fence illusion” : gaps of silence dramatically distort perception of a sentence, while bursts of noise do not

39. Next Time: Taste Smell Touch Balance