IEICE Trans

Atsunori OGAWA

Multi-Source Domain Generalization Using Domain Attributes for Recurrent Neural Network Language Models Naohiro TAWARA Atsunori OGAWA Tomoharu IWATA Hiroto ASHIKAWA Tetsunori KOBAYASHI Tetsuji OGAWA
Publication: Publication Date: 2022/01/01 Vol. E105-D No. 1 pp. 150-160 Type of Manuscript: PAPER Category: Natural Language Processing Keyword: domain attribute, mixture-of-experts, recurrent adaptive mixture model, domain generalization, language model,
	Summary \| Full Text:PDF(4MB)

Feature Based Domain Adaptation for Neural Network Language Models with Factorised Hidden Layers Michael HENTSCHEL Marc DELCROIX Atsunori OGAWA Tomoharu IWATA Tomohiro NAKATANI
Publication: Publication Date: 2019/03/01 Vol. E102-D No. 3 pp. 598-608 Type of Manuscript: PAPER Category: Speech and Hearing Keyword: language model, LSTM, domain adaptation, unsupervised, latent Dirichlet allocation,
	Summary \| Full Text:PDF(1.7MB)

Automatic Vocabulary Adaptation Based on Semantic and Acoustic Similarities Shoko YAMAHATA Yoshikazu YAMAGUCHI Atsunori OGAWA Hirokazu MASATAKI Osamu YOSHIOKA Satoshi TAKAHASHI
Publication: IEICE TRANSACTIONS on Information and Systems Publication Date: 2014/06/01 Vol. E97-D No. 6 pp. 1488-1496 Type of Manuscript: Special Section PAPER (Special Section on Advances in Modeling for Real-world Speech Information Processing and its Application) Category: Speech Recognition Keyword: out-of-vocabulary, vocabulary adaptation, semantic similarity, acoustic similarity,
	Summary \| Full Text:PDF(1.1MB)

Efficient Combination of Likelihood Recycling and Batch Calculation for Fast Acoustic Likelihood Calculation Atsunori OGAWA Satoshi TAKAHASHI Atsushi NAKAMURA
Publication: IEICE TRANSACTIONS on Information and Systems Publication Date: 2011/03/01 Vol. E94-D No. 3 pp. 648-658 Type of Manuscript: PAPER Category: Speech and Hearing Keyword: fast acoustic likelihood calculation, state likelihood recycling, batch state likelihood calculation, combined acceleration technique, acoustic backing-off,
	Summary \| Full Text:PDF(823.2KB)