Illinois Speech Day May 10, 2010 TTI-Chicago, Rooms 526/530 6045 S. Kenwood Ave., Chicago Parking information Contact: Karen Livescu, klivescu at uchicago.edu UIUC contact: Mark Hasegawa-Johnson Northwestern contact: Janet Pierrehumbert Speech Day will be a day-long meeting consisting of presentations and discussion on the theme of computational models of speech. Presenters will be faculty, postdocs, and students at U. Chicago, Northwestern, UIUC, and TTI-C. Presentations will include completed and ongoing work, with the goal of fostering interaction among the sites. Please contact Karen Livescu to register. Space is limited.

---

Preliminary schedule (see also complete list of attendees)

• 9:30-10:30 light breakfast


• 10:30-10:50 Jennifer Cole, UIUC (Prosody and reduction in spontaneous speech)
• 10:50-11:10 Jont Allen, UIUC (Identification and conformation of acoustic speech cues, and the role of conflicting cues on consonant confusions)
• 11:10-11:30 Anjali Menon, UIUC (Verification of speech cues for stops and fricatives)
• 11:30-11:50 Panying Rong and Chris Carignan, UIUC (Lingual co-articulation of nasalized vowels in American English)
• 11:50-12:10 Ryan Shosted, UIUC (The nasal specification of oral obstruents)
• 12:10-12:30 Stephen Levinson, UIUC (Language acquisition on the iCub humanopid robot)


• 12:30-1:30 lunch (catered)


• 1:30-1:50 Matt Goldrick, Northwestern (Mixture model analysis of phonetic variation in speech errors )
• 1:50-2:10 Charlotte Vaughn, Northwestern (Classification of orthographically transcribed speech by native language using character n-grams)
• 2:10-2:30 Jordana Heller, Northwestern (Influences of word topicality on the predictability of repeated words during reading)
• 2:30-2:50 Karen Livescu, TTI-C (Recent work on articulatory modeling for speech recognition)
• 2:50-3:10 Joseph Keshet, TTI-C (Direct Loss Minimization for Forced Alignment)


• 3:10-4:40 coffee break/posters:

Sam Bowman, U. Chicago (Context-dependent articulatory feature modeling)
Woojae Han, UIUC (Quantifying speech-loss and consonant-dependence in hearing-impaired listeners)
Lisa Hesterberg, Northwestern (Distribution of the Arabic plural and some OCP effects on the plural)
Chi Hu, UIUC (Finite State ASR Dictionary using Gesture Pattern Vectors as Units)
Jui-Ting Huang, UIUC (Prosody-Dependent Acoustic Modeling Using Variable-Parameter Hidden Markov Models)
Po-Sen Huang, UIUC (Synchrony and Asynchrony Modeling for Audio-Visual Event Detection)
Arthur Kantor, UIUC (Smoothing posteriors lowers the frame error rate in speech recognition)
John Labiak, U. Chicago (Nearest-neighbor phonetic classification)
Anna Margolis, U. Washington/TTI-C (Cross-domain dialogue act tagging)
Sravana Reddy, U. Chicago (Refining pronunciation lexicons using recognition errors)
Riya Singh, UIUC (How do humans with cochlear lesions understand speech?)
Siwei Wang, U. Chicago (Tone Recognition with Sequential Graphical Models)


• 4:40-5:00 Jason Riggle, U. Chicago (Initial steps toward modeling and recognition of American Sign Language: Measures of articulatory complexity)
• 5:00-5:20 Max Bane, U. Chicago (Longitudinal phonetic variation in a closed system)
• 5:20-5:40 Morgan Sonderegger, U. Chicago (Automatic discriminative measurement of voice onset time)


• 6:30- go out for dinner