1 / 68

Speech Recognition

Speech Recognition. Speech Sounds of American English. Introduction. Speech was created since the inception of the human race . In contrast writing is at most a few thousand years old.

faye
Download Presentation

Speech Recognition

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Speech Recognition Speech Sounds of American English

  2. Introduction • Speech was created since the inception of the human race. • In contrast writing is at most a few thousand years old. • Speech is available to anyone and everyone learns to speak without formal instruction and attains a comparable level of skill and fluency. • Speech is the most common and most natural manifestation of language. • Phonetics, the study of speech sounds, is the bedrock of the scientific study of language. Henderson in 1877 said that “The form of language is its sounds”. Veton Këpuska

  3. Introduction • “Language is not a cultural artifact that we learn the way we learn to tell time or how the federal government works. Instead, it is a distinct piece of biological makeup of our brain.” – Steven Pinker “The Language Instinct: How the Mind Creates Language” Veton Këpuska

  4. Introduction • Language is: • a complex, • specialized skill, • which develops in the child spontaneously, • without conscious effort or formal instruction, • is deployed without awareness of its underlying logic, • is qualitatively the same in every individual, and • is distinct from more general abilities to process information or behave intelligently. • For these reasons some cognitive scientist have described language as a psychological faculty, a mental organ, a neural system, and a computational module. • The term “instinct” is preferred to all the above. Veton Këpuska

  5. Speech Sounds of American English • There are over 40 speech sounds in American English which can be organized by their basic manner of production Veton Këpuska

  6. Speech Sounds of American English • Vowels, glides, and consonants differ in degree of constriction • Sonorant • nasals, • liquids and • glides) • are consonants that have no pressure build up at constriction • Nasal consonants that have no pressure build up at constriction • Continuantconsonants that do not block airflow in oral cavity Veton Këpuska

  7. Phonemes of American English Veton Këpuska

  8. Phonetic Alphabets Reference Veton Këpuska

  9. Phonetic Alphabets Reference Veton Këpuska

  10. ARPA-Bet Phone Set • SPHINX\ARPAbetExample.pdf Veton Këpuska

  11. SPHYNX Phone Set Veton Këpuska

  12. Vocal Tract Veton Këpuska

  13. Vowel Production • No significant constriction in the vocal tract • Usually produced with periodic excitation • Acoustic characteristics depend on the position of the jaw, tongue, and lips Veton Këpuska

  14. Vowels of American English • There are approximately 18 vowels in American English made up of monothongs, diphthongs, and reduced vowels (schwa’s) • They are often described by the articulatory features: High/Low, Front/Back, Retroflexed, Rounded, and Tense/Lax Veton Këpuska

  15. Articulatory Features • Place of articulation: • Palatal <-> Velar • Palatal: Front • Velar: Back • Manner of Articulation: • Degree of Closure • Secondary Articulations: e.g. Lip Rounding • Aperture: • Close – the narrowest constriction • Open – widest opening of vocal track. • Mid – midway opening relative to close position and open position. • Vocal Track Shape • Raising or lowering the tongue • Advancing or retracting the body of the tongue. • Raising or lowering the jaw • Rounding the lips. Veton Këpuska

  16. Vowels Veton Këpuska

  17. Spectrograms of the Cardinal Vowels Veton Këpuska

  18. Vowel Formant Averages • Vowels are often characterized by the lower three formants: • High/Low is correlated with the first formant, F1 • Front/Back is correlated with the second formant, F2 • Retroflexion is marked by a low third formant, F3 Veton Këpuska

  19. Vowel Durations • Each vowel has a different intrinsic duration • Schwa’s have distinctly shorter durations (50ms) • /I, ℇ, Λ, Ʊ/ are the shortest monothongs • Context can greatly influence vowel duration Veton Këpuska

  20. Happy Little Vowel Chart Robb's "So inaccurate, yet so useful." Veton Këpuska

  21. fricatives Veton Këpuska

  22. Fricative Production • Turbulence produced at narrow constriction • Constriction position determines acoustic characteristics • Can be produced with periodic excitation Veton Këpuska

  23. Fricatives of American English • There are 8 fricatives in American English • Four places of articulation: Labio-Dental (Labial), Interdental (Dental), Alveolar, and Palato-Alveolar (Palatal) • They are often described by the features Voiced/Unvoiced, or Strident/Non-Strident (constriction behind alveolar ridge) Veton Këpuska

  24. Spectrograms of Unvoiced Fricatives Veton Këpuska

  25. Fricative Energy • Strident fricatives tend to be stronger than non-strident fricatives. Veton Këpuska

  26. Fricative Durations • Voiced fricatives tend to be shorter than unvoiced fricatives. Veton Këpuska

  27. Examples of Fricative Voicing Contrast Veton Këpuska

  28. Friendly Little Consonant Chart Robb's "Somewhat more accurate, yet somewhat less useful." Veton Këpuska

  29. What is this word? Veton Këpuska

  30. STOPS Veton Këpuska

  31. Stop Production • Complete closure in the vocal tract, pressure build up • Sudden release of the constriction, turbulence noise • Can have periodic excitation during closure Veton Këpuska

  32. Stops of American English • There are 6 stop consonants in American English • Three places of articulation: Labial, Alveolar, and Velar • Each place of articulation has a voiced and unvoiced stop • Unvoiced stops are typically aspirated • Voiced stops usually exhibit a “voice-bar’’ during closure • Information about formant transitions and release useful for classification Veton Këpuska

  33. Spectrograms of Unvoiced Stops Veton Këpuska

  34. Examples of Stop Voicing Contrast Veton Këpuska

  35. Singleton Stop Durations Veton Këpuska

  36. Voicing Cues for Stops Veton Këpuska

  37. /s/-Stop Durations Veton Këpuska

  38. Examples of Front and Back Velars Veton Këpuska

  39. What is this word? Veton Këpuska

  40. NASALS Veton Këpuska

  41. Nasal Production • Velum lowering results in airflow through nasal cavity • Consonants produced with closure in oral cavity • Nasal murmurs have similar spectral characteristics Veton Këpuska

  42. Nasal of American English • Three places of articulation: Labial, Alveolar, and Velar • Nasal consonants are always attached to a vowel, though can form an entire syllable in unstressed environments ([n], [m], [ŋ]) • /ŋ/ is always post-vocalic in English • Place identified by neighboring formant transitions Veton Këpuska

  43. Spectrograms of Nasals Veton Këpuska

  44. What is this word? Veton Këpuska

  45. semivowels Veton Këpuska

  46. Semivowel Production • Constriction in vocal tract, no turbulence • Slower articulatory motion than other consonants • Laterals form complete closure with tongue tip, airflow via sides of constriction Veton Këpuska

  47. Semivowels of American English • There are 4 semivowels in American English • Sometimes referred to as Liquids or Glides • Glides are a more extreme articulation of a corresponding vowel • Similar, though more extreme, formant positions • Generally weaker due to narrower constriction • Semivowels are always attached to a vowel, though /l/ can form an entire syllable in unstressed environments ([l]) Veton Këpuska

  48. Spectrograms of Semivowels Veton Këpuska

  49. Acoustic Properties of Semivowels • /w/ and /l/ are the most confusable semivowels • /w/ is characterized by a very low F1, F2 • Typically a rapid spectral fallow above F2 • /l/ is characterized by a low F1 and F2 • Often presence of high frequency energy • Postvocalic /l/ characterized by minimal spectral discontinuity, gradual motion of formants • /y/ is characterized by very low F1, very high F2 • /y/ only occurs in a syllable onset position (i.e., pre-vocalic) • /r/ is characterized by a very low F3 • Prevocalic F3 < medial F3 < postvocalic F3 Veton Këpuska

  50. What is this word? Veton Këpuska

More Related