Zobrazeno 1 - 10
of 152
pro vyhledávání: '"Thomas, Hain"'
Publikováno v:
Journal of Big Data, Vol 8, Iss 1, Pp 1-16 (2021)
Abstract Speech based human-machine interaction and natural language understanding applications have seen a rapid development and wide adoption over the last few decades. This has led to a proliferation of studies that investigate Error detection and
Externí odkaz:
https://doaj.org/article/2a752c4a8ebc49488211eb508c5b87fc
Publikováno v:
Frontiers in Signal Processing, Vol 2 (2022)
Separation of speech mixtures in noisy and reverberant environments remains a challenging task for state-of-the-art speech separation systems. Time-domain audio speech separation networks (TasNets) are among the most commonly used network architectur
Externí odkaz:
https://doaj.org/article/02f17f0adaa14f3cbd21b78e7075c7b9
Publikováno v:
Interspeech 2022.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::fb9abf851f2b9524795d101c6134c9bf
https://doi.org/10.21437/interspeech.2022-10182
https://doi.org/10.21437/interspeech.2022-10182
Publikováno v:
Interspeech 2022.
Multilingual speech recognition has drawn significant attention as an effective way to compensate data scarcity for low-resource languages. End-to-end (e2e) modelling is preferred over conventional hybrid systems, mainly because of no lexicon require
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f1c2dc1e15fcb225625a6fbd8e171302
https://doi.org/10.21437/interspeech.2022-11449
https://doi.org/10.21437/interspeech.2022-11449
Publikováno v:
2022 International Workshop on Acoustic Signal Enhancement (IWAENC).
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::ec75f49b9afad96dbbd68847f52b4170
https://doi.org/10.1109/iwaenc53105.2022.9914752
https://doi.org/10.1109/iwaenc53105.2022.9914752
This paper proposes an unsupervised data selection method by using a submodular function based on contrastive loss ratios of target and training data sets. A model using a contrastive loss function is trained on both sets. Then the ratio of frame-lev
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5c469d023f7760bb389ca51cd287eb64
http://arxiv.org/abs/2207.12028
http://arxiv.org/abs/2207.12028
Publikováno v:
Journal of Big Data, Vol 8, Iss 1, Pp 1-16 (2021)
Speech based human-machine interaction and natural language understanding applications have seen a rapid development and wide adoption over the last few decades. This has led to a proliferation of studies that investigate Error detection and classifi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::05ea7785fc71a4309fa665cd3b42a650
https://doaj.org/article/2a752c4a8ebc49488211eb508c5b87fc
https://doaj.org/article/2a752c4a8ebc49488211eb508c5b87fc
Autor:
Jose Antonio Lopez Saenz, Thomas Hain
Publikováno v:
ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a1cbd68e59837001694811c978aefea2
https://doi.org/10.1109/icassp43922.2022.9746720
https://doi.org/10.1109/icassp43922.2022.9746720
Separation of speech mixtures in noisy and reverberant environments remains a challenging task for state-of-the-art speech separation systems. Time-domain audio speech separation networks (TasNets) are among the most commonly used network architectur
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::128c8be9a2e3d04ff8c989b89630da43
https://eprints.whiterose.ac.uk/186716/1/frsip-02-856968.pdf
https://eprints.whiterose.ac.uk/186716/1/frsip-02-856968.pdf
Training of speech enhancement systems often does not incorporate knowledge of human perception and thus can lead to unnatural sounding results. Incorporating psychoacoustically motivated speech perception metrics as part of model training via a pred
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::db3add9d3ca4e0ed48f0223300380ff5
http://arxiv.org/abs/2203.12369
http://arxiv.org/abs/2203.12369