Using Phonologically Weighted Levenshtein Distances for the Prediction of Microscopic Intelligibility

Abstract : This article presents a new method for analyzing Automatic Speech Recognition (ASR) results at the phonological feature level. To this end the Levenshtein distance algorithm is refined in order to take into account the distinctive features opposing substituted phonemes. This method allows to survey features additions or deletions, providing microscopic qualitative information as a complement to word recognition scores. To explore the relevance of the qualitative data gathered by this method, a study is conducted on a speech corpus simulating presbycusis effects on speech perception at eight severity stages. Consonantic features additions and deletions in ASR outputs are analyzed and put in relation with intelligibility data collected in 30 human subjects. ASR results show monotonic trends in most conso- nantic features along the degradation conditions, which appear to be consistent with the misperceptions that could be observed in human subjects.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01474904
Contributor : Open Archive Toulouse Archive Ouverte (oatao) <>
Submitted on : Thursday, February 23, 2017 - 11:52:58 AM
Last modification on : Thursday, June 27, 2019 - 4:27:51 PM
Long-term archiving on : Wednesday, May 24, 2017 - 1:12:37 PM

File

fontan_17158.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01474904, version 1
  • OATAO : 17158

Collections

Citation

Lionel Fontan, Isabelle Ferrané, Jérôme Farinas, Julien Pinquier, Xavier Aumont. Using Phonologically Weighted Levenshtein Distances for the Prediction of Microscopic Intelligibility. Annual conference Interspeech (INTERSPEECH 2016), Sep 2016, San Francisco, CA, United States. pp. 650-654. ⟨hal-01474904⟩

Share

Metrics

Record views

95

Files downloads

626