Week 9

Lecture 18 October (Tuesday, in Sed):
Chatbot models, NLU and ASR

Presentation : PDF | ODP

Recordings: sorry, the room microphone did not work

Mandatory reading

Jurafsky and Martin, Speech and Language Processing, 3. ed. (edition of Jan. 2022)

• Ch. 24, "Dialogue systems and chatbots"
  • Sections 24.2 & 24.3
• Ch. 25, "Phonetics"
  • Sections 25.1-25.5 (excluding the details not discussed in class)
• Chap 26, "Speech Recognition and ASR"
  • Sections 26.1 and 26.5 (excluding the part on statistical significance)

Lab session October 20 (Thursday, in Smalltalk):

1. Exercise: Analyse the following dialogue (audio, transcript) and answer the following questions:
   • How are the dialogue turns structured, based on the observed linguistic cues?
   • What kind of speech acts are used through the dialogue, according to Searle's taxonomy?
   • What are the grounding signals and strategies used through the dialogue?
   • Can you find some examples of conversational implicatures?
   • List a few (2-3) deictic markers occurring in the dialogue.
   • Find 4-5 disfluencies in the dialogue and analyse them according to Shriberg's model. 
2. Introduction to NumPy (loosely based on this tutorial)

Lab session October 25:

The lab session will cover the following exercises (extracted from previous exam questions).