FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3160200310> ?p ?o ?g. }

• W3160200310 abstract "This study addresses unsupervised subword modeling, i.e., learning acoustic feature representations that can distinguish between subword units of a language. We propose a two-stage learning framework that combines self-supervised learning and cross-lingual knowledge transfer. The framework consists of autoregressive predictive coding (APC) as the front-end and a cross-lingual deep neural network (DNN) as the back-end. Experiments on the ABX subword discriminability task conducted with the Libri-light and ZeroSpeech 2017 databases showed that our approach is competitive or superior to state-of-the-art studies. Comprehensive and systematic analyses at the phoneme- and articulatory feature (AF)-level showed that our approach was better at capturing diphthong than monophthong vowel information, while also differences in the amount of information captured for different types of consonants were observed. Moreover, a positive correlation was found between the effectiveness of the back-end in capturing a phoneme's information and the quality of the cross-lingual phone labels assigned to the phoneme. The AF-level analysis together with t-SNE visualization results showed that the proposed approach is better than MFCC and APC features in capturing manner and place of articulation information, vowel height, and backness information. Taken together, the analyses showed that the two stages in our approach are both effective in capturing phoneme and AF information. Nevertheless, monophthong vowel information is less well captured than consonant information, which suggests that future research should focus on improving capturing monophthong vowel information." @default.
• W3160200310 created "2021-05-24" @default.
• W3160200310 creator A5048860892 @default.
• W3160200310 creator A5053373302 @default.
• W3160200310 date "2021-01-01" @default.
• W3160200310 modified "2023-10-16" @default.
• W3160200310 title "The Effectiveness of Unsupervised Subword Modeling With Autoregressive and Cross-Lingual Phone-Aware Networks" @default.
• W3160200310 cites W1494198834 @default.
• W3160200310 cites W1524333225 @default.
• W3160200310 cites W1557247526 @default.
• W3160200310 cites W1631260214 @default.
• W3160200310 cites W1778492285 @default.
• W3160200310 cites W1975728937 @default.
• W3160200310 cites W1984076147 @default.
• W3160200310 cites W2020607164 @default.
• W3160200310 cites W2025722797 @default.
• W3160200310 cites W2085628288 @default.
• W3160200310 cites W2100768664 @default.
• W3160200310 cites W2102113734 @default.
• W3160200310 cites W2128032727 @default.
• W3160200310 cites W2147768505 @default.
• W3160200310 cites W2160815625 @default.
• W3160200310 cites W2187089797 @default.
• W3160200310 cites W2293634267 @default.
• W3160200310 cites W2327501763 @default.
• W3160200310 cites W2399576818 @default.
• W3160200310 cites W2407080277 @default.
• W3160200310 cites W2513125788 @default.
• W3160200310 cites W2514741789 @default.
• W3160200310 cites W2516890051 @default.
• W3160200310 cites W2594951208 @default.
• W3160200310 cites W2641832364 @default.
• W3160200310 cites W2748598007 @default.
• W3160200310 cites W2759889345 @default.
• W3160200310 cites W2785415724 @default.
• W3160200310 cites W2785860501 @default.
• W3160200310 cites W2786608204 @default.
• W3160200310 cites W2787223168 @default.
• W3160200310 cites W2787426069 @default.
• W3160200310 cites W2787447541 @default.
• W3160200310 cites W2826003142 @default.
• W3160200310 cites W2842511635 @default.
• W3160200310 cites W2889228998 @default.
• W3160200310 cites W2895297209 @default.
• W3160200310 cites W2905332681 @default.
• W3160200310 cites W2927673779 @default.
• W3160200310 cites W2949510815 @default.
• W3160200310 cites W2950414763 @default.
• W3160200310 cites W2962693497 @default.
• W3160200310 cites W2962824366 @default.
• W3160200310 cites W2963618559 @default.
• W3160200310 cites W2963620343 @default.
• W3160200310 cites W2963799213 @default.
• W3160200310 cites W2964121744 @default.
• W3160200310 cites W2964245029 @default.
• W3160200310 cites W2972374322 @default.
• W3160200310 cites W2972841524 @default.
• W3160200310 cites W2972943112 @default.
• W3160200310 cites W2973026522 @default.
• W3160200310 cites W2973157397 @default.
• W3160200310 cites W2995181338 @default.
• W3160200310 cites W29952999 @default.
• W3160200310 cites W3005578234 @default.
• W3160200310 cites W3016181583 @default.
• W3160200310 cites W3044483536 @default.
• W3160200310 cites W3093096176 @default.
• W3160200310 cites W3095361818 @default.
• W3160200310 cites W3096216486 @default.
• W3160200310 cites W3096359985 @default.
• W3160200310 cites W3098361150 @default.
• W3160200310 cites W3104842308 @default.
• W3160200310 cites W3105242324 @default.
• W3160200310 cites W3125709657 @default.
• W3160200310 cites W3134136786 @default.
• W3160200310 cites W3144810982 @default.
• W3160200310 cites W3150635893 @default.
• W3160200310 doi "https://doi.org/10.1109/ojsp.2021.3076914" @default.
• W3160200310 hasPublicationYear "2021" @default.
• W3160200310 type Work @default.
• W3160200310 sameAs 3160200310 @default.
• W3160200310 citedByCount "0" @default.
• W3160200310 crossrefType "journal-article" @default.
• W3160200310 hasAuthorship W3160200310A5048860892 @default.
• W3160200310 hasAuthorship W3160200310A5053373302 @default.
• W3160200310 hasBestOaLocation W31602003101 @default.
• W3160200310 hasConcept C108583219 @default.
• W3160200310 hasConcept C120665830 @default.
• W3160200310 hasConcept C121332964 @default.
• W3160200310 hasConcept C138885662 @default.
• W3160200310 hasConcept C149782125 @default.
• W3160200310 hasConcept C154945302 @default.
• W3160200310 hasConcept C159877910 @default.
• W3160200310 hasConcept C192209626 @default.
• W3160200310 hasConcept C204321447 @default.
• W3160200310 hasConcept C2776401178 @default.
• W3160200310 hasConcept C2778707766 @default.
• W3160200310 hasConcept C2779581591 @default.
• W3160200310 hasConcept C28490314 @default.
• W3160200310 hasConcept C33923547 @default.
• W3160200310 hasConcept C41008148 @default.

• next