Weakly Supervised Multi-Embeddings Learning of Acoustic Models

Weakly Supervised Multi-Embeddings Learning of Acoustic Models. Synnaeve, G. & Dupoux, E. April, 2015. arXiv:1412.6645 [cs]

Paper doi abstract bibtex

We trained a Siamese network with multi-task same/different information on a speech dataset, and found that it was possible to share a network for both tasks without a loss in performance. The first task was to discriminate between two same or different words, and the second was to discriminate between two same or different talkers.

@misc{synnaeve_weakly_2015,
	title = {Weakly {Supervised} {Multi}-{Embeddings} {Learning} of {Acoustic} {Models}},
	url = {http://arxiv.org/abs/1412.6645},
	doi = {10.48550/arXiv.1412.6645},
	abstract = {We trained a Siamese network with multi-task same/different information on a speech dataset, and found that it was possible to share a network for both tasks without a loss in performance. The first task was to discriminate between two same or different words, and the second was to discriminate between two same or different talkers.},
	urldate = {2023-01-09},
	publisher = {arXiv},
	author = {Synnaeve, Gabriel and Dupoux, Emmanuel},
	month = apr,
	year = {2015},
	note = {arXiv:1412.6645 [cs]},
	keywords = {Computer Science - Computation and Language, Computer Science - Machine Learning, Computer Science - Sound, I.2.6, I.2.7, I.5.1},
}

Downloads: 0

{"_id":"8Ky4hm7Qbd627vE2Z","bibbaseid":"synnaeve-dupoux-weaklysupervisedmultiembeddingslearningofacousticmodels-2015","author_short":["Synnaeve, G.","Dupoux, E."],"bibdata":{"bibtype":"misc","type":"misc","title":"Weakly Supervised Multi-Embeddings Learning of Acoustic Models","url":"http://arxiv.org/abs/1412.6645","doi":"10.48550/arXiv.1412.6645","abstract":"We trained a Siamese network with multi-task same/different information on a speech dataset, and found that it was possible to share a network for both tasks without a loss in performance. The first task was to discriminate between two same or different words, and the second was to discriminate between two same or different talkers.","urldate":"2023-01-09","publisher":"arXiv","author":[{"propositions":[],"lastnames":["Synnaeve"],"firstnames":["Gabriel"],"suffixes":[]},{"propositions":[],"lastnames":["Dupoux"],"firstnames":["Emmanuel"],"suffixes":[]}],"month":"April","year":"2015","note":"arXiv:1412.6645 [cs]","keywords":"Computer Science - Computation and Language, Computer Science - Machine Learning, Computer Science - Sound, I.2.6, I.2.7, I.5.1","bibtex":"@misc{synnaeve_weakly_2015,\n\ttitle = {Weakly {Supervised} {Multi}-{Embeddings} {Learning} of {Acoustic} {Models}},\n\turl = {http://arxiv.org/abs/1412.6645},\n\tdoi = {10.48550/arXiv.1412.6645},\n\tabstract = {We trained a Siamese network with multi-task same/different information on a speech dataset, and found that it was possible to share a network for both tasks without a loss in performance. The first task was to discriminate between two same or different words, and the second was to discriminate between two same or different talkers.},\n\turldate = {2023-01-09},\n\tpublisher = {arXiv},\n\tauthor = {Synnaeve, Gabriel and Dupoux, Emmanuel},\n\tmonth = apr,\n\tyear = {2015},\n\tnote = {arXiv:1412.6645 [cs]},\n\tkeywords = {Computer Science - Computation and Language, Computer Science - Machine Learning, Computer Science - Sound, I.2.6, I.2.7, I.5.1},\n}\n\n","author_short":["Synnaeve, G.","Dupoux, E."],"key":"synnaeve_weakly_2015","id":"synnaeve_weakly_2015","bibbaseid":"synnaeve-dupoux-weaklysupervisedmultiembeddingslearningofacousticmodels-2015","role":"author","urls":{"Paper":"http://arxiv.org/abs/1412.6645"},"keyword":["Computer Science - Computation and Language","Computer Science - Machine Learning","Computer Science - Sound","I.2.6","I.2.7","I.5.1"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"misc","biburl":"https://bibbase.org/zotero/mxmplx","dataSources":["aXmRAq63YsH7a3ufx"],"keywords":["computer science - computation and language","computer science - machine learning","computer science - sound","i.2.6","i.2.7","i.5.1"],"search_terms":["weakly","supervised","multi","embeddings","learning","acoustic","models","synnaeve","dupoux"],"title":"Weakly Supervised Multi-Embeddings Learning of Acoustic Models","year":2015}