Speech repains, intonational phrases, and discourse markers: modeling speakers’ utterances in spoken dialogue

By Peter A. Heeman, James F. Allen. 1999

Subfields: Dialogue & Conversational AI, Speech & Audio

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Dialogue & Conversational AI

1. Mari Ostendorf
2. Brian Roark

In Speech & Audio

1. Mari Ostendorf
2. Andreas Stolcke
3. Brian Roark
4. Elizabeth Shriberg

Citations in the corpus

• Enriching speech recognition with automatic detection of sentence boundaries and disfluencies (2006, background): “Heeman and Allen [2] proposed a statistical language model to identify part-of-speech (POS) tags, discourse markers, speech repairs, and intonational phrases.”
• Robust garden path parsing (2004, methodology): “There have been other approaches to modeling the joint probability of disfluency labels and words, including Stolcke, Shriberg, Hakkani-T¨ur and T¨ur (1999) and Heeman and Allen (1999).”
• Exploiting prosodic breaks in language modeling with random forests (2008, unknown): “This joint model is immediately usable if our goal of ASR is the simultaneous recognition of words and prosodic breaks [20, 5]:”