The article “Speaker-adapted confidence measures for speech recognition of video lectures“, by Isaías Sánchez-Cortina and other MLLP members, has been accepted for publication in the May 2016 volume of Elsevier’s Computer Speech & Language journal, and is already available online. Mr Sánchez-Cortina will also be defending his PhD thesis this Friday at Universitat Politècnica de València.
The article’s authors, MLLP members Isaías Sánchez Cortina, Jesús Andrés Ferrer, Albert Sanchis and Alfons Juan, have summarized it with the following highlights:
- A new, particular logistic regression model is proposed to improve confidence measures for automatic speech recognition.
- Speaker-adapted models are proposed to further improve confidence measures.
- Empirical results are provided showing that speaker-adapted models outperform their non-adapted counterparts.
- The improvement of confidence measures is shown to be useful on an interactive speech transcription application.
Elsevier’s Computer Speech & Language, an official publication of the International Speech Communication Association (ISCA), publishes reports of original research related to the recognition, understanding, production, coding and mining of speech and language, encouraging an interdisciplinary approach to speech and language research and technology. With a JCR 2014 impact factor of 1.753 (5-year: 1.789), it’s a Q2 journal in the category of Computer Science: Artificial Intelligence.
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.
In related news, Isaías Sánchez Cortina, first author of this new article and former MLLP member, will be defending this week his PhD thesis on “Confidence Measures for Automatic and Interactive Speech Recognition” (advisors: Dr Alfons Juan Ciscar and Dr Alberto Sanchis); you can read here the dissertation’s abstract. The public PhD defence will take place next Friday, 15 January 2016, at 12:00 PM, at the Universitat Politècnica de València’s Dept. of Computer Systems and Computation (DSIC).