Zobrazeno 1 - 7
of 7
pro vyhledávání: '"Jason Riesa"'
Much of text-to-speech research relies on human evaluation, which incurs heavy costs and slows down the development process. The problem is particularly acute in heavily multilingual applications, where recruiting and polling judges can take weeks. W
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b3c2078509e36e897d48e0ce5cd4abb0
Autor:
Alexis Conneau, Ankur Bapna, Yu Zhang, Min Ma, Patrick von Platen, Anton Lozhkov, Colin Cherry, Ye Jia, Clara Rivera, Mihir Kale, Daan van Esch, Vera Axelrod, Simran Khanuja, Jonathan Clark, Orhan Firat, Michael Auli, Sebastian Ruder, Jason Riesa, Melvin Johnson
We introduce XTREME-S, a new benchmark to evaluate universal cross-lingual speech representations in many languages. XTREME-S covers four task families: speech recognition, classification, speech-to-text translation and retrieval. Covering 102 langua
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7d5112276248b7b6fa2427d9ed857587
Publikováno v:
EMNLP (1)
State-of-the-art multilingual models depend on vocabularies that cover all of the languages the model will expect to see at inference time, but the standard methods for generating those vocabularies are not ideal for massively multilingual applicatio
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::623fc1d9776966a7676ce9268cb1ba06
Publikováno v:
EMNLP/IJCNLP (1)
We propose a practical scheme to train a single multilingual sequence labeling model that yields state of the art results and is small and fast enough to run on a single CPU. Starting from a public multilingual BERT checkpoint, our final model is 6x
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b3443548f1b256b12a1dff0a589f2e5c
https://doi.org/10.18653/v1/d19-1374
https://doi.org/10.18653/v1/d19-1374
Autor:
Aditya Siddhant, Naveen Ari, Karthik Raman, Melvin Johnson, Jason Riesa, Ankur Bapna, Henry Tsai, Orhan Firat
Publikováno v:
AAAI
The recently proposed massively multilingual neural machine translation (NMT) system has been shown to be capable of translating over 100 languages to and from English within a single model (Aharoni, Johnson, and Firat 2019). Its improved translation
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c25424f2bbee8d412632ba2763404141
Publikováno v:
EMNLP
We address fine-grained multilingual language identification: providing a language code for every token in a sentence, including codemixed text containing multiple languages. Such text is prevalent online, in documents, social media, and message boar
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6c693be5937a7a66cec1f0e151ef44a5
http://arxiv.org/abs/1810.04142
http://arxiv.org/abs/1810.04142
Publikováno v:
Interspeech 2006.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::08294a79e8795ac976917e3166393d71
https://doi.org/10.21437/interspeech.2006-261
https://doi.org/10.21437/interspeech.2006-261