New MLLP article on computer-aided transcription of video lectures published in Elsevier’s Speech Communication

MLLPThe article Efficiency and usability study of innovative computer-aided transcription strategies for video lecture repositories“, by Juan Daniel Valor Miró and other MLLP members, has been published in the November 2015 volume of Elsevier’s Speech Communication journal.

The article’s authors, MLLP members Juan Daniel Valor Miró, Joan Albert Silvestre-Cerdà, Jorge Civera and Alfons Juan, together with Universitat Politècnica de València’s Carlos Turró, have summarized it with the following highlights:

  • Real-life evaluation of automatic transcriptions in a large video lecture repository.
  • Three different evaluation protocols were compared to minimise user supervision time.
  • Discovered dependencies between transcription quality and time spent in supervision.
  • Attained up to 70% time reduction with post-editing, as compared to transcribing from scratch.
  • Lecturers valued simplicity over sophisticated but more efficient protocols.

Elsevier’s Speech Communication, a publication of the European Association for Signal Processing (EURASIP) and the International Speech Communication Association (ISCA), is an interdisciplinary journal whose primary objective is to fulfil the need for the rapid dissemination and thorough discussion of basic and applied research results. With a JCR 2014 impact factor of 1.256 (5-year: 1.786), it’s a Q2 journal in the category of Acoustics and a Q3 journal in the category of Computer Science: Interdisciplinary Applications.

This new article follows a line of publications by Juan Daniel Valor Miró in journals and conferences on the evaluation of the post-edition of automatic transcriptions and translations of video lectures, including a 2014 article in Routledge’s Open Learning: The Journal of Open, Distance and e-Learning titled “Evaluating intelligent interfaces for post-editing automatic transcriptions of online video lectures“, and (for more recent results) a 2015 paper for EC-TEL 2015 titled “Efficient Generation of High-Quality Multilingual Subtitles for Video Lecture Repositories“.

In the last six years (2009-2015), MLLP members have published over 15 international journal articles (Pattern Recognition Letters, 51, 2015; Pattern Recognition, 45 (9), 2012; IEEE Trans. Audio Speech Lang. Process., 20 (2), 2012; …) and over 50 international conference papers (AMTA 2014; ICASSP 2013;  EACL 2012; ICDAR 2011; …). You can browse through all of the group’s 200+ publications (starting from 1991) in the Publications section in our website.

Posted in All, Publications and tagged , , , , , .