learn.ifs.name

Phonetics and Speech Feature Extraction

Jurafsky & Martin, Speech and Language Processing, 3rd ed. draft, ch. 15

18 cards in “15.1 Phonetic transcription” · whole deck

15.1 Phonetic transcription

  1. What is a phone?
    A speech sound, represented with a symbol adapted from the Roman alphabet. A word's pronunciation is represented as a string of phones.
  2. What is the IPA and when was it first developed?
    The International Phonetic Alphabet: the evolving standard for transcribing the world's languages, first developed in 1888.
  3. What is the ARPAbet?
    A simple phonetic alphabet (Shoup, 1980) using ASCII symbols to represent an American-English subset of the IPA.
  4. Is English orthography transparent or opaque with respect to phones? Give an example.
    Relatively opaque: one letter can represent very different sounds. c is [k] in cougar but [s] in cell. Spanish is much more transparent.
  5. List spellings (besides k) that can represent the phone [k] in English.
    c (cougar), x (fox [f aa k s]), ck (jackal), cc (raccoon).
  6. CELEX dictionary: size and contents?
    English pronunciations for 160,595 wordforms, with syllabification, stress, and morphological and part-of-speech information.
  7. CMU Pronouncing Dictionary: size and license?
    Open-source; about 134,000 wordforms.
  8. UNISYN dictionary: size and special feature?
    About 110,000 words, fine-grained; gives syllabifications, stress, and pronunciations for dozens of English dialects. Free for research.
  9. Which languages does the LDC distribute pronunciation lexicons for?
    Egyptian Arabic, Dutch, English, German, Japanese, Korean, Mandarin, Spanish.
  10. What is a phonetically annotated corpus?
    A collection of waveforms hand-labeled with the corresponding string of phones.
  11. TIMIT corpus: who built it and what does it contain?
    Texas Instruments, MIT and SRI. 6300 read sentences: 10 sentences each from 630 speakers, drawn from a set of 2342 sentences (some chosen for dialect shibboleths, others to maximize diphone coverage).
  12. How was TIMIT phonetically labeled?
    Hand-labeled phone sequences were automatically aligned with the wavefile, and the automatic phone boundaries were then manually hand-corrected.
  13. What is a time-aligned transcription?
    A transcription in which each phone is associated with a start and end time in the waveform.
  14. Which narrow-transcription phenomena does TIMIT mark (e.g. in she had your dark suit in greasy wash water)?
    Palatalization of [d] in had, unreleased final stop in dark, glottalization of final [t] in suit to [q], and the flap of [t] in water.
  15. Switchboard Transcription Project: size and alignment level?
    3.5 hours of sentences from the Switchboard corpus, time-aligned at the syllable level.
  16. What phenomena appear in the Switchboard transcription of they're kind of in between right now?
    Vowel reduction (they're, of), coda deletion (kind, right), and resyllabification (the [v] of of becomes the onset of in).
  17. Buckeye corpus: what is it?
    A phonetically transcribed corpus of spontaneous American speech: about 300,000 words from 40 talkers.
  18. Name phonetically transcribed corpora for German and Mandarin.
    The Kiel corpus (German); Mandarin corpora from the Chinese Academy of Social Sciences.

Source