Matches in SemOpenAlex for { <https://semopenalex.org/work/W2528542014> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W2528542014 abstract "This Cross-lingual knowledge sharing based acoustic modeling methods are usually used in Automatic Speech Recognition (ASR) of languages which do not have enough transcribed speech for acoustic model (AM) training. Conventional methods such as IPA based universal acoustic modeling have been proved to be effective under matched acoustic conditions, while usually poorly preformed when mismatch appears between the target language and the source languages. This paper proposes a method of multi-lingual unsupervised AM training for zero-resourced languages under mismatch conditions. The proposed method includes two main steps. In the first step, initial AM of the target low-resourced language was obtained using multi-task training method, in which original source language data and mapped source language data are jointly used. In the second step, AM of the target language is trained using automatically transcribed target language data, in the way of iteratively training new AMs and adapting the initial AMs. Experiments were conducted on a corpus with 100 hours untranscribed Japanese speech and 300 hours transcribed speech of other languages. The best result achieved by this paper is 51.75% character error rate (CER), which obtains 24.78% absolute reduction compared to baseline IPA system." @default.
- W2528542014 created "2016-10-14" @default.
- W2528542014 creator A5039631808 @default.
- W2528542014 creator A5060165452 @default.
- W2528542014 creator A5061055212 @default.
- W2528542014 date "2016-06-01" @default.
- W2528542014 modified "2023-09-23" @default.
- W2528542014 title "Multi-lingual unsupervised acoustic modeling using multi-task deep neural network under mismatch conditions" @default.
- W2528542014 cites W139772320 @default.
- W2528542014 cites W1524333225 @default.
- W2528542014 cites W1606350000 @default.
- W2528542014 cites W1965842648 @default.
- W2528542014 cites W1975759754 @default.
- W2528542014 cites W1984985727 @default.
- W2528542014 cites W1993660824 @default.
- W2528542014 cites W1993952617 @default.
- W2528542014 cites W2025198378 @default.
- W2528542014 cites W2026369565 @default.
- W2528542014 cites W2040903810 @default.
- W2528542014 cites W2056786202 @default.
- W2528542014 cites W2088275353 @default.
- W2528542014 cites W2091746061 @default.
- W2528542014 cites W2114214391 @default.
- W2528542014 cites W2133267619 @default.
- W2528542014 cites W2155667629 @default.
- W2528542014 cites W2160815625 @default.
- W2528542014 cites W2164505566 @default.
- W2528542014 cites W2189391786 @default.
- W2528542014 cites W2402741009 @default.
- W2528542014 cites W2537062360 @default.
- W2528542014 cites W3148201686 @default.
- W2528542014 cites W55333121 @default.
- W2528542014 cites W66627554 @default.
- W2528542014 cites W84880661 @default.
- W2528542014 doi "https://doi.org/10.1109/iccsn.2016.7586635" @default.
- W2528542014 hasPublicationYear "2016" @default.
- W2528542014 type Work @default.
- W2528542014 sameAs 2528542014 @default.
- W2528542014 citedByCount "0" @default.
- W2528542014 crossrefType "proceedings-article" @default.
- W2528542014 hasAuthorship W2528542014A5039631808 @default.
- W2528542014 hasAuthorship W2528542014A5060165452 @default.
- W2528542014 hasAuthorship W2528542014A5061055212 @default.
- W2528542014 hasConcept C137293760 @default.
- W2528542014 hasConcept C154945302 @default.
- W2528542014 hasConcept C155635449 @default.
- W2528542014 hasConcept C162324750 @default.
- W2528542014 hasConcept C187736073 @default.
- W2528542014 hasConcept C204321447 @default.
- W2528542014 hasConcept C2780451532 @default.
- W2528542014 hasConcept C28490314 @default.
- W2528542014 hasConcept C40969351 @default.
- W2528542014 hasConcept C41008148 @default.
- W2528542014 hasConcept C50644808 @default.
- W2528542014 hasConcept C61328038 @default.
- W2528542014 hasConceptScore W2528542014C137293760 @default.
- W2528542014 hasConceptScore W2528542014C154945302 @default.
- W2528542014 hasConceptScore W2528542014C155635449 @default.
- W2528542014 hasConceptScore W2528542014C162324750 @default.
- W2528542014 hasConceptScore W2528542014C187736073 @default.
- W2528542014 hasConceptScore W2528542014C204321447 @default.
- W2528542014 hasConceptScore W2528542014C2780451532 @default.
- W2528542014 hasConceptScore W2528542014C28490314 @default.
- W2528542014 hasConceptScore W2528542014C40969351 @default.
- W2528542014 hasConceptScore W2528542014C41008148 @default.
- W2528542014 hasConceptScore W2528542014C50644808 @default.
- W2528542014 hasConceptScore W2528542014C61328038 @default.
- W2528542014 hasLocation W25285420141 @default.
- W2528542014 hasOpenAccess W2528542014 @default.
- W2528542014 hasPrimaryLocation W25285420141 @default.
- W2528542014 hasRelatedWork W1591482254 @default.
- W2528542014 hasRelatedWork W1892788530 @default.
- W2528542014 hasRelatedWork W1993660824 @default.
- W2528542014 hasRelatedWork W2008737763 @default.
- W2528542014 hasRelatedWork W2037564206 @default.
- W2528542014 hasRelatedWork W2040903810 @default.
- W2528542014 hasRelatedWork W2059019626 @default.
- W2528542014 hasRelatedWork W2107553890 @default.
- W2528542014 hasRelatedWork W2147590749 @default.
- W2528542014 hasRelatedWork W2333651221 @default.
- W2528542014 hasRelatedWork W2463237750 @default.
- W2528542014 hasRelatedWork W2607725416 @default.
- W2528542014 hasRelatedWork W2773832794 @default.
- W2528542014 hasRelatedWork W2890244912 @default.
- W2528542014 hasRelatedWork W2971766686 @default.
- W2528542014 hasRelatedWork W3011926625 @default.
- W2528542014 hasRelatedWork W3037057938 @default.
- W2528542014 hasRelatedWork W3046125265 @default.
- W2528542014 hasRelatedWork W3104842308 @default.
- W2528542014 hasRelatedWork W7946549 @default.
- W2528542014 isParatext "false" @default.
- W2528542014 isRetracted "false" @default.
- W2528542014 magId "2528542014" @default.
- W2528542014 workType "article" @default.