HCI.TOP
C7.04

Accent, Dialect, and Code-Switching

Cards in this group · 7

  1. C7.04.1Training-data distribution determines whose speech is recognized well
  2. C7.04.2Recognition degrades on mixed Chinese–English and dialect words
  3. C7.04.3Accuracy gaps exclude particular populations
  4. C7.04.4Mid-utterance language switches require live language identification
  5. C7.04.5A wrong language decision forces the rest of the sentence through the wrong phonology
  6. C7.04.6Dialects share vocabulary with the standard but differ in pronunciation, so they are misread as near-homophones
  7. C7.04.7Accent and dialect gains depend on training data from those speakers, not on algorithms alone