Skip to content

Detecting Errors in Parsing of Speech Transcript primarily due to Active Linguistic Repair

Notifications You must be signed in to change notification settings

sayarghoshroy/Detect_Parse_Errors_due_to_Linguistic_Repair

Repository files navigation

Detecting Errors in Parsing of Speech Transcript primarily due to Active Linguistic Repair

Open In Colab

Analyze Transcript of NPTEL Lectures

  • Standard English Language tokenized text separated into sentences by newlines
  • Six text files used for analysis: Course 101101058, Lectures 14 to 19
  • Detect errors in Dependency Parse output by Spacy's parser
  • Capture Instances of Active Linguistic Repair
  • Find cases where Repair leads to a failed Parse

Refer to the Jupyter Notebook for a proper understanding of the Implementation

Refer to the full report for:

  • Task Definition
  • Comments on the Parser
  • Explanation on Violating Conditions
  • Analysis of Sentences from our Observation Set
  • Linguistic Repair
  • How occurence of Repair affects parse output
  • Various Modules
  • Algorithms Used
  • Implementation Details
  • Related Work

The Presentation Slides can be viewed here

About

Detecting Errors in Parsing of Speech Transcript primarily due to Active Linguistic Repair

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published