Digital Systems: Hardware Organization and Design

Digital Systems: Hardware Organization and Design
11/10/2018 Speech Recognition Speech Sounds of American English Architecture of a Respresentative 32 Bit Processor

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 a 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”. 10 November 2018 Veton Këpuska

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”’ 10 November 2018 Veton Këpuska

Introduction “Back in the thirteenth century, the German King, Frederick II, conducted a diabolical experiment intended to discover what language children would naturally grow up to speak if never spoken to.” “So King Frederick took babies from their mothers at birth and placed them in the care of nurses who were forbidden to speak in their presence.” “An additional rule was imposed as well: the nurses were not allowed to touch the infants.” 10 November 2018 Veton Këpuska

Introduction To his great dismay, Frederick’s experiment was cut short, but not before something tragically significant regarding human nature was revealed. As you may have guessed, the babies grew up to speak no language at all because they all died!” In the year 1248, an Italian historian (Salimbene di Adam) recorded: “They could not live without petting.” The babies literally died for want of touch. 10 November 2018 Veton Këpuska

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. 10 November 2018 Veton Këpuska

Speech Sounds of American English
10 November 2018 Veton Këpuska

Digital Systems: Hardware Organization and Design 11/10/2018 Speech Sounds of American English There are over 40 speech sounds in American English which can be organized by their basic manner of production Manner Class Number Vowel 18 Fricatives 8 Stops 6 Nasals 3 Semivowels 4 Affricates 2 Aspirant 1 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

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 Continuant consonants that do not block airflow in oral cavity 10 November 2018 Veton Këpuska

Phonemes of American English
Digital Systems: Hardware Organization and Design 11/10/2018 Phonemes of American English 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Phonetic Alphabets Reference

ARPA-Bet Phone Set SPHINX\ARPAbetExample.pdf 10 November 2018
Veton Këpuska

SPHYNX Phone Set 11/10/2018 Phoneme Example Translation 1. AA odd AA D 2. AE at AE T 3. AH hut HH AH T 4. AO ought AO T 5. AW cow K AW 6. AY hide HH AY D 7. B be B IY 8. CH cheese CH IY Z 9. D dee D IY 10. DH thee DH IY 11. EH Ed EH D 12. ER hurt HH ER T 13. EY ate EY T 14. F fee F IY 15. G green G R IY N 16. HH he HH IY 17. JH gee JH IY 18. IH it IH T 19. IY eat IY T 20. Phoneme Example Translation 21. K key K IY 22. L lee L IY 23. M me M IY 24. N knee N IY 25. NG ping P IH NG 26. OW oat OW T 27. OY toy T OY 28. P pee P IY 29. R read R IY D 30. S sea S IY 31. SH she SH IY 32. T tea T IY 33. TH theta TH EY T AH 34. UH hood HH UH D 35. UW two T UW 36. V vee V IY 37. W we W IY 38. Y yield Y IY L D 39. Z zee Z IY 40. ZH seizure S IY ZH ER 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

American English Phonemes
Digital Systems: Hardware Organization and Design 11/10/2018 American English Phonemes Vowels (Voiced) Fricatives (Voiced & Unvoiced) Stops (Voiced & Unvoiced) Nasals (Voiced) Semivowels (Voiced) Affricates (Fricative with Voicing) Aspirants (Voiced or Unvoiced) 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Vocal Tract 10 November 2018 Veton Këpuska

Vowels 10 November 2018 Veton Këpuska

Vocal Cords/Folds 10 November 2018 Veton Këpuska

Vocal Fold Airflow 10 November 2018 Veton Këpuska

11/10/2018 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, e.g.: 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Vowels of American English
Digital Systems: Hardware Organization and Design 11/10/2018 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 (tounge, lips, etc.): High/Low, Front/Back, Retroflexed, Rounded, and Tense/Lax 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

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 trackt. 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. 10 November 2018 Veton Këpuska

Spectrogram – what it is?
A spectrogram is a visual representation of a sound. It displays the amplitude of the frequency components of the signal over time. sound spectrogram a graphic representation, produced by a sound spectrograph, of the frequency, intensity, duration, and variation with time of the resonance of a sound or series of sounds. Dictionary.com Unabridged Based on the Random House Dictionary, © Random House, Inc 10 November 2018 Veton Këpuska

Spectrograms of the Cardinal Vowels
Digital Systems: Hardware Organization and Design 11/10/2018 Spectrograms of the Cardinal Vowels 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Vowel Formant Averages
Digital Systems: Hardware Organization and Design 11/10/2018 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 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

11/10/2018 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 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Happy Little Vowel Chart
Digital Systems: Hardware Organization and Design 11/10/2018 Happy Little Vowel Chart Robb's "So inaccurate, yet so useful." 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

A Slightly Different Vowel Chart.

A Slightly Different Vowel Chart.
IPA English Vowel Chart 10 November 2018 Veton Këpuska

Vowel Place of Articulation

fricatives 10 November 2018 Veton Këpuska

11/10/2018 Fricative Production Turbulence produced at narrow constriction Position of the constriction determines acoustic characteristics Can be produced with periodic excitation 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Fricatives of American English
Digital Systems: Hardware Organization and Design 11/10/2018 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) 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Spectrograms of Unvoiced Fricatives
Digital Systems: Hardware Organization and Design 11/10/2018 Spectrograms of Unvoiced Fricatives 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

11/10/2018 Fricative Energy Strident fricatives tend to be stronger than non-strident fricatives. 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

11/10/2018 Fricative Durations Voiced fricatives tend to be shorter than unvoiced fricatives. 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Examples of Fricative Voicing Contrast
Digital Systems: Hardware Organization and Design 11/10/2018 Examples of Fricative Voicing Contrast 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Friendly Little Consonant Chart
Digital Systems: Hardware Organization and Design 11/10/2018 Robb's Friendly Little Consonant Chart "Somewhat more accurate, yet somewhat less useful." 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

11/10/2018 What is this word? 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Help? 10 November 2018 Veton Këpuska

STOPS 10 November 2018 Veton Këpuska

11/10/2018 Stop Production Complete closure in the vocal tract, pressure build up Sudden release of the constriction, turbulence noise Can have periodic excitation during closure 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Stops of American English
Digital Systems: Hardware Organization and Design 11/10/2018 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 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Spectrograms of Unvoiced Stops
Digital Systems: Hardware Organization and Design 11/10/2018 Spectrograms of Unvoiced Stops 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Examples of Stop Voicing Contrast
Digital Systems: Hardware Organization and Design 11/10/2018 Examples of Stop Voicing Contrast 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Singleton Stop Durations
Digital Systems: Hardware Organization and Design 11/10/2018 Singleton Stop Durations 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

11/10/2018 Voicing Cues for Stops 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

11/10/2018 /s/-Stop Durations 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Examples of Front and Back Velars
Digital Systems: Hardware Organization and Design 11/10/2018 Examples of Front and Back Velars 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Help? 10 November 2018 Veton Këpuska

NASALS 10 November 2018 Veton Këpuska

11/10/2018 Nasal Production Velum lowering results in airflow through nasal cavity Consonants produced with closure in oral cavity Nasal murmurs have similar spectral characteristics 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Nasal of American English
Digital Systems: Hardware Organization and Design 11/10/2018 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 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Spectrograms of Nasals
Digital Systems: Hardware Organization and Design 11/10/2018 Spectrograms of Nasals 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

semivowels 10 November 2018 Veton Këpuska

11/10/2018 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 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Semivowels of American English
Digital Systems: Hardware Organization and Design 11/10/2018 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]) 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Spectrograms of Semivowels
Digital Systems: Hardware Organization and Design 11/10/2018 Spectrograms of Semivowels 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Acoustic Properties of Semivowels
Digital Systems: Hardware Organization and Design 11/10/2018 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 change 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 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Affricates 10 November 2018 Veton Këpuska

11/10/2018 Affricate Production There are two affricates in American English: Alveolar-stop palatal-fricative pairs Sudden release of the constriction, turbulence noise Can have periodic excitation during closure 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

aspirant 10 November 2018 Veton Këpuska

11/10/2018 Aspirant Production There is one aspirant in American English: /h/ (e.g., “hat’’) Produced by generating turbulence excitation at glottis No constriction in the vocal tract, normal formant excitation Sub-glottal coupling results in little energy in F1 region Periodic excitation can be present in medial position 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Spectrograms of Affricates and Aspirant
Digital Systems: Hardware Organization and Design 11/10/2018 Spectrograms of Affricates and Aspirant 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Additional Constrains

Phonotactic Constraints
Digital Systems: Hardware Organization and Design 11/10/2018 Phonotactic Constraints Phonotactics is the study of allowable sound sequences Analyses of word-initial and -final clusters reveal: 73 distinct initial clusters (about 10 “foreign’’ clusters) 208 distinct final clusters Can be used to eliminate impossible phoneme sequences: /tk/ can’t end a word, and /kt/ can’t begin a word, Therefore, */… t k t …/ is an impossible sequence 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Word-Initial Consonants from MWP Dictionary
Digital Systems: Hardware Organization and Design 11/10/2018 Word-Initial Consonants from MWP Dictionary 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Syllable 10 November 2018 Veton Këpuska

Word Origin & History A unit of pronunciation having one vowel sound, with or without surrounding consonants, forming the whole or a part of a word; Example: there are two syllables in water and three in inferno. Syllable comes via Old French and Latin from Greek sullabē, from sun- ‘together’ and lambanein ‘take’. A syllable is basically a group of sounds ‘taken together’ and uttered with a single effort. 10 November 2018 Veton Këpuska

Merriam Webster Definition
1: a unit of spoken language that is next bigger than a speech sound (e.g., phoneme) and consists of one or more vowel sounds alone or of a syllabic consonant alone or of either with one or more consonant sounds preceding or following 2: one or more letters (as syl, la, and ble) in a word (as syl*la*ble) usually set off from the rest of the word by a centered dot or a hyphen and roughly corresponding to the syllables of spoken language and treated as helps to pronunciation or as guides to placing hyphens at the end of a line 3: the smallest conceivable expression or unit of something 10 November 2018 Veton Këpuska

Dictionary Definition
Noun an uninterrupted segment of speech consisting of a vowel sound, a diphthong or a syllabic consonant, with or without preceding or following consonant sounds: “eye,” “sty,” “act,” and “should” are English words of one syllable. “eyelet,” “stifle,” “enact,” and “shouldn't” are two-syllable words one or more written letters or characters representing more or less exactly such an element of speech. the slightest portion or amount of speech or writing; the least mention: Do not breathe a syllable of all this. verb (used with object) to utter in syllables; articulate. to represent by syllables. to utter syllables; speak. 10 November 2018 Veton Këpuska

A syllable is a unit of organization for a sequence of speech sounds.
For example, the word water is composed of two syllables: wa and ter. A syllable is typically made up of a syllable nucleus (most often a vowel) with optional initial and final margins (typically, consonants). Syllables are often considered the phonological "building blocks" of words. They can influence the rhythm of a language, its prosody, its poetic meter and its stress patterns. 10 November 2018 Veton Këpuska

Syllabic writing began several hundred years before the first letters
Syllabic writing began several hundred years before the first letters. The earliest recorded syllables are on tablets written around 2800 BC in the Sumerian city of Ur. This shift from pictograms to syllables has been called "the most important advance in the history of writing".[1] A word that consists of a single syllable (like English dog) is called a monosyllable (and is said to be monosyllabic). Similar terms include: disyllable (and disyllabic) for a word of two syllables; trisyllable (and trisyllabic) for a word of three syllables; and polysyllable (and polysyllabic), which may refer either to a word of more than three syllables or to any word of more than one syllable. 10 November 2018 Veton Këpuska

Sonority and Syllables
Syllables are generally organized around a phonological property called sonority. basically: sonority = perceived loudness Sonorants (vowels, liquids, nasals, glides) have lots of sonority; obstruents (stops and fricatives) have less. Basic idea: the most sonorous segments in a syllable form the “peak” or nucleus of the syllable. vowels make good peaks; sonorant consonants are second-best; obstruents are really bad… 10 November 2018 Veton Këpuska

Examples [bæd] as in bed: is a well-formed syllable in English.
[æ] high sonority [b] [d] low sonority 10 November 2018 Veton Këpuska

[blænd] as in blend: works well, too. [æ] high sonority [l] [n]
[b] [d] low sonority 10 November 2018 Veton Këpuska

Technical Terms [æ] high sonority [l] [n] [b] [d] low sonority
SONORITY PEAK 10 November 2018 Veton Këpuska

Technical Terms Sonority Peak form the nucleus of the syllable
[æ] high sonority [l] [n] [b] [d] low sonority nucleus 10 November 2018 Veton Këpuska

The sonority peak forms the nucleus of the syllable. The sounds that precede the nucleus form the syllable onset. onset 10 November 2018 Veton Këpuska

The sonority peak forms the nucleus of the syllable. The sounds that precede the nucleus form the syllable onset. The sounds that follow the nucleus form the syllable coda. coda 10 November 2018 Veton Këpuska

Syllable Formation In order to figure out how to organize a word into syllables, first identify the syllable nuclei = vowels (V) and any syllabic consonants (C) Example: “America” [A M E R I K A] | | | | | | | V C V C V C V Then identify any potential onsets to each syllable = consonants preceding the nuclei [A M E R I K A] s s s s 10 November 2018 Veton Këpuska

11/10/2018 The Syllable Syllable structure captures many useful generalizations Phoneme realization often depends on syllabification Many phonological rules depend on syllable structure Syllable structure is predicated on the notion of ranking the speech sounds in terms of their sonority values 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Sonority The Sonority Sequencing Principle (SSP)[1][2] is a phonotactic principle that aims to outline the structure of a syllable in terms of sonority. The SSP states that the center of a syllable, namely the syllable nucleus, often a vowel, constitutes a sonority peak that is preceded and/or followed by a sequence of segments – consonants – with progressively decreasing sonority values (i.e., the sonority has to fall toward both edges of the syllable). The sonority values of segments are determined by a sonority hierarchy. 10 November 2018 Veton Këpuska

t, which is a stop, the lowest on the sonority scale; next is
A good example for the SSP in English is the one-syllable word "trust": The first consonant in the syllable onset is t, which is a stop, the lowest on the sonority scale; next is r, a liquid which is more sonorous, then we have the vowel u /ʌ/ - the sonority peak; next, in the syllable coda, is s, a fricative, and last is another stop, t. 10 November 2018 Veton Këpuska

Some languages possess syllables that violate the SSP (Russian and English, for example) while other languages strictly adhere to it, even requiring larger intervals on the sonority scale: In Italian for example, a syllable-initial stop must be followed by either a liquid, a glide or a vowel, but not by a fricative (except: [ps] borrowed words like: pseudonimo, psicologia). Some languages allow a sonority "plateau"; that is, two adjacent tautosyllabic consonants with the same sonority level. 10 November 2018 Veton Këpuska

Syllables and Sonority
Digital Systems: Hardware Organization and Design 11/10/2018 Syllables and Sonority Utterances can be divided into syllables The number of syllables equals the number of sonority peaks Within any syllable, there is a segment constituting a sonority peak that is preceded and/or followed by a sequence of segments with progressively decreasing sonority values 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

11/10/2018 The Syllable Template Branches marked by “o” are optional Nucleus must contain a non-obstruent Sonority decreases away from nucleus Affix contains only coronals: Only the last syllable in a word can have an affix /sp/, /st/, and /sk/ are treated as single obstruents 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

11/10/2018 Some Examples 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Words Containing /r/ and /l/
Digital Systems: Hardware Organization and Design 11/10/2018 Words Containing /r/ and /l/ 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Acoustic Realizations of /r/
Digital Systems: Hardware Organization and Design 11/10/2018 Acoustic Realizations of /r/ 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Acoustic Realizations of /l/
Digital Systems: Hardware Organization and Design 11/10/2018 Acoustic Realizations of /l/ 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Allophonic Variations at Syllable Boundaries
Digital Systems: Hardware Organization and Design 11/10/2018 Allophonic Variations at Syllable Boundaries 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

11/10/2018 References “Acoustics of American English Speech – A Dynamic Approach”, J. Olive, A. Greenwood, and J. Coleman, Springer-Verlag 1993. “Articulatory-Acoustic-Auditory Relationships”, Kenneth Stevens, in The Handbook of Phonetic Sciences, Ed. William Hardcastle and John Laver, Blackwell Publishers, 1997. 10 November 2018 Veton Këpuska Architecture of a Respresentative 32 Bit Processor

Digital Systems: Hardware Organization and Design

Similar presentations

Presentation on theme: "Digital Systems: Hardware Organization and Design"— Presentation transcript:

Similar presentations

About project

Feedback

Log in

Auth with social network:

Digital Systems: Hardware Organization and Design

Similar presentations

Presentation on theme: "Digital Systems: Hardware Organization and Design"— Presentation transcript:

Similar presentations

About project

Feedback