Extracting semantically-coherent keyphrases from speech

  • Diana Inkpen Sch. of Info. Technology and Eng., University of Ottawa, 800 King Edward St., Ottawa, Ont. K1N6H5, Canada
  • Alain Désilets Institute for Information Technology, National Research Council, 1200 Montreal Rd., Ottawa, Ont. K1AOR6, Canada
Keywords: Algorithms, Audio systems, Broadcasting, Error analysis, Semantics, Speech, Pointwise mutual information (PMI), Semantic coherence, Speech extractor systems, Word error rates (WER)


A method for extracting keyphrase from spoken audio is discussed. The approach is based on the extractor system developed for text by Turney. The extractor uses a supervised learning approach to maximize the overlap between machine extracted and human extracted keyphrases. Each key word is detected with low similarity to its closest semantic neighbor, using an algorithm.
How to Cite
Inkpen D, Désilets A. Extracting semantically-coherent keyphrases from speech. Canadian Acoustics [Internet]. 2004Sep.1 [cited 2019Oct.16];32(3):130-1. Available from: https://jcaa.caa-aca.ca/index.php/jcaa/article/view/1662
Proceedings of the Acoustics Week in Canada