Prosody and Research (Guest Speaker Chris Heffner, 5/21/13)

Chris Heffner, a NSF Graduate Research Fellow in the Program in Neuroscience and Cognitive Science at UMD, spoke about prosody and about the undergraduate research process.

Slideshow available here.

NOTES (thanks to Megan Chao):

  • IPA system, phonetics, and different sounds

    • Jay asked about Chinese sounds on the IPA system

    • X-sound is “c with a loopy thing”

    • Keep in mind that it is an international system, so it works for Chinese, too

  • Lots of transcription ambiguity

    • Punctuation

    • Reading IPA sounds

      • No spaces between words

  • PROSODY – the ~variations~

    • Suprasegmental traits

    • Also, there are TREES

    • Not quite like syntactic trees, can’t just take part of the tree and know what it means?

    • Various way to transcribe

  • Sound pressure graphs of speech

    • How to segment the graph and decode into words

    • Amplitudes correspond to going up and down really fast (like vocal cords)

      • Bigger = louder

    • People don’t actually pause in between every word

      • Where do they stop/start?

  • Throw things into the Prosody Box and don’t worry about it

  • Preserve proximal context and slow down distal context

    • John said he would obey a rebel.

    • Slowed context causes people to report hearing fewer words in a sentence

    • Proximal cuts should be weighted more heavily than distal cues

  • Research is a long and arduous journey, kids (917 days)

  • Make money! Mechanical Turks (sp?)

  • Phoneticians are more likely to have an extra fold in brain

    • Which came first? The phonetician or the fold?

Leave a Reply

Your email address will not be published. Required fields are marked *