Zobrazeno 1 - 10
of 26
pro vyhledávání: '"Eugen Beck"'
In this work, we show that a factored hybrid hidden Markov model (FH-HMM) which is defined without any phonetic state-tying outperforms a state-of-the-art hybrid HMM. The factored hybrid HMM provides a link to transducer models in the way it models p
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::aa92b5643e3b4324719d420080d1592b
Publikováno v:
INTERSPEECH
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::fa73a89c9b2c045664321da861ad3a66
https://doi.org/10.21437/interspeech.2020-1164
https://doi.org/10.21437/interspeech.2020-1164
Publikováno v:
INTERSPEECH
Interspeech 2020
Interspeech 2020
Phoneme-based acoustic modeling of large vocabulary automatic speech recognition takes advantage of phoneme context. The large number of context-dependent (CD) phonemes and their highly varying statistics require tying or smoothing to enable robust t
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5a0ce7eef31675d70bb703a3a62bbbe7
https://doi.org/10.21437/interspeech.2020-1244
https://doi.org/10.21437/interspeech.2020-1244
Publikováno v:
ICASSP
ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Despite advances in neural language modeling, obtaining a good model on a large scale multi-domain dataset still remains a difficult task. We propose training methods for building neural language models for such a task, which are not only domain robu
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::1df966fc403350f909aa99a6d15bbd6e
https://doi.org/10.1109/icassp40776.2020.9054399
https://doi.org/10.1109/icassp40776.2020.9054399
Autor:
Markus Kitza, Albert Zeyer, Ralf Schlüter, Hermann Ney, Eugen Beck, Christoph Lüscher, Wilfried Michel, Kazuki Irie
Publikováno v:
INTERSPEECH
Interspeech 2019
Interspeech 2019
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d9d20f297b7d6e038334eed4269dcc61
https://doi.org/10.21437/interspeech.2019-1780
https://doi.org/10.21437/interspeech.2019-1780
Publikováno v:
INTERSPEECH
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::e5efb532ee7a5fd589f66fd4df3ddbca
https://doi.org/10.21437/interspeech.2019-1817
https://doi.org/10.21437/interspeech.2019-1817
Publikováno v:
IEEE transactions on pattern analysis and machine intelligence : TPAMI 41(2), 502-514 (2019). doi:10.1109/TPAMI.2017.2788434
IEEE Transactions on Pattern Analysis and Machine Intelligence
IEEE Transactions on Pattern Analysis and Machine Intelligence
IEEE transactions on pattern analysis and machine intelligence : TPAMI 41(2), 502-514 (2019). doi:10.1109/TPAMI.2017.2788434
Published by IEEE, New York, NY
Published by IEEE, New York, NY
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f3ad2d758a1651a92aa15d1215287377
https://publications.rwth-aachen.de/record/755689
https://publications.rwth-aachen.de/record/755689
Publikováno v:
INTERSPEECH
Interspeech 2017
Baixas, France : International Speech Communication Association, ISCA 944-948 (2017). doi:10.21437/Interspeech.2017-1073
Interspeech 2017 : Stockholm, Sweden, 20-24 August 2017 / chair: Francisco Lacerda
Interspeech 2017 : Stockholm, Sweden, 20-24 August 2017 / chair: Francisco Lacerda18. Annual Conference of the International Speech Communication Association, Interspeech 2017, Stockholm, Sweden, 2017-08-20-2017-08-24
Interspeech 2017
Baixas, France : International Speech Communication Association, ISCA 944-948 (2017). doi:10.21437/Interspeech.2017-1073
Interspeech 2017 : Stockholm, Sweden, 20-24 August 2017 / chair: Francisco Lacerda
Interspeech 2017 : Stockholm, Sweden, 20-24 August 2017 / chair: Francisco Lacerda18. Annual Conference of the International Speech Communication Association, Interspeech 2017, Stockholm, Sweden, 2017-08-20-2017-08-24
[Interspeech 2017, 2017-08-20 - 2017-08-20, Stockholm, Sweden] Interspeech 2017, Stockholm, Sweden, 20 Aug 2017 - 20 Aug 2017; Stockholm 944-948 (2017). doi:10.21437/Interspeech.2017-1073
Published by Stockholm
Published by Stockholm
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::181bf327e03a6a1d11f7efcdaa838b5a
https://doi.org/10.21437/interspeech.2017-1073
https://doi.org/10.21437/interspeech.2017-1073
Publikováno v:
Speech and Computer ISBN: 9783319664286
SPECOM
SPECOM
In this paper we describe the RWTH Aachen keyword search (KWS) system developed in the course of the IARPA Babel program. We put focus on acoustic modeling with neural networks and evaluate the full pipeline with respect to the KWS performance. At th
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::6e41be3a13539f4f33aa72a7e01100c1
https://doi.org/10.1007/978-3-319-66429-3_72
https://doi.org/10.1007/978-3-319-66429-3_72
Publikováno v:
INTERSPEECH
Scopus-Elsevier
Scopus-Elsevier
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0cdb80edbdc08a2eb71604d92d592f94
https://doi.org/10.21437/interspeech.2015-320
https://doi.org/10.21437/interspeech.2015-320