Publication Details
Unsupervised Word Segmentation from Speech with Attention
Boito Marcely Z. (INRIA)
Ondel Yang Lucas Antoine Francois, Mgr., Ph.D. (DCGM FIT BUT)
Berard Alexandre (INRIA)
Yvon Francois (LIMSI)
Villavicencio Aline (UESSEX)
Besacier Laurent (UGA)
computational language documentation,encoder-decoder models, attentional models, unsupervised word segmentation.
We present a first attempt to perform attentional word segmentation directly from the speech signal, with the final goal to automatically identify lexical units in a low-resource, unwritten language (UL). Our methodology assumes a pairing between recordings in the UL with translations in a well-resourced language. It uses Acoustic Unit Discovery (AUD) to convert speech into a sequence of pseudo-phones that is segmented using neural soft-alignments produced by a neural machine translation model. Evaluation uses an actual Bantu UL, Mboshi; comparisons to monolingual and bilingual baselines illustrate the potential of attentional word segmentation for language documentation.
@INPROCEEDINGS{FITPUB12242, author = "Pierre Godard and Z. Marcely Boito and Francois Antoine Lucas Yang Ondel and Alexandre Berard and Francois Yvon and Aline Villavicencio and Laurent Besacier", title = "Unsupervised Word Segmentation from Speech with Attention", pages = "2678--2682", booktitle = "Proceeding of Interspeech 2018", journal = "Proceedings of Interspeech - on-line", volume = 2018, number = 9, year = 2018, location = "Hyderabad, IN", publisher = "International Speech Communication Association", ISSN = "1990-9772", doi = "10.21437/Interspeech.2018-1308", language = "english", url = "https://www.fit.vut.cz/research/publication/12242" }