WP4 – Intelligent interaction with users

What is Intelligent Interaction?

Intelligent interaction departs from two premises. Firstly, that fully-automated approaches cannot yet be relied on to produce fully usable texts and so user supervision must be somehow incorporated into the transcription/translation workflow. Secondly, that this interaction needs to be intelligent in order for it to be sustainable over vast collections of video material. So, rather than having the user post-edit entire texts, the transLectures system will identify those transcription/translation segments it is least confident about and direct the user to them for their input. These corrections will then be fed into the system in real-time, essentially allowing it to learn from past mistakes in a process of ongoing improvement.

Video: Watch the transLectures intelligent interaction demo in our Progress section!

Objectives

The main goal of this work package is to demonstrate that accurate transcriptions and translations can be obtained via a process of intelligent user interaction. This objective presupposes the initial generation of transcriptions and translations of an acceptable quality. At transLectures we will also be exploring different ways of interacting with the system, to reflect the multiple roles of potential users.

Deliverables

You can find information and download links for the transLectures tools in our Tools section.

Milestones

  • Initial tools for intelligent interaction (M6)
  • First improvement of tools for intelligent interaction (M12)
  • Second improvement of tools for intelligent interaction (M24)
  • Third improvement of tools for intelligent interaction (M30)
  • Final tools for intelligent interaction (M36)

Click here for the full list of transLectures public reports and deliverables published to date.