eduzhai > Applied Sciences > Engineering >

GIPFA Generating IPA Pronunciation from Audio

  • Save

... pages left unread,continue reading

Document pages: 8 pages

Abstract: Transcribing spoken audio samples into International Phonetic Alphabet (IPA)has long been reserved for experts. In this study, we instead examined the useof an Artificial Neural Network (ANN) model to automatically extract the IPApronunciation of a word based on its audio pronunciation, hence its nameGenerating IPA Pronunciation From Audio (GIPFA). Based on the French Wikimediadictionary, we trained our model which then correctly predicted 75 of the IPApronunciations tested. Interestingly, by studying inference errors, the modelmade it possible to highlight possible errors in the dataset as well asidentifying the closest phonemes in French.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...