ISCA Archive SpeechProsody 2010
ISCA Archive SpeechProsody 2010

Enriching speech engineering

Shrikanth Narayanan

Engineering approaches offer a rich set of possibilities for facilitating fundamental advances in speech prosody research. Likewise, theories and models of speech prosody continue to guide the creation of engineering techniques that attempt to capture the rich tapestry of information contained in speech. Together, this symbiosis is leading to the development of a variety of technology applications that are enriched by linguistic and paralinguistic prosodic information. This talk will highlight some recent examples on enriching--and the enriched--speech engineering. We will first illustrate instrumental technologies such as real time MRI and motion capture that are helping illuminate the intricate speech production details valuable in understanding prosodic structure. Next, we will discuss some of the computational tools for characterizing and modeling prosody. We will conclude with a discussion of technology applications, such as speech-to-speech translation and behavioral informatics, that exploit such enriched information.

Cite as: Narayanan, S. (2010) Enriching speech engineering. Proc. Speech Prosody 2010, paper 1001

  author={Shrikanth Narayanan},
  title={{Enriching speech engineering}},
  booktitle={Proc. Speech Prosody 2010},
  pages={paper 1001}