Zobrazeno 1 - 5
of 5
pro vyhledávání: '"Felix de Chaumont Quitry"'
Publikováno v:
IEEE Signal Processing Letters. 27:630-634
The deployment of deep networks on mobile devices requires to efficiently use the scarce computational resources, expressed as either available memory or computing cost. When addressing multiple tasks simultaneously, it is extremely important to shar
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a95ced211e743968d5f839dba1fc2bda
https://doi.org/10.1109/lsp.2020.2988158
https://doi.org/10.1109/lsp.2020.2988158
Publikováno v:
IEEE Signal Processing Letters. 27:600-604
We explore self-supervision as a way to learn general purpose audio representations. Specifically, we propose two self-supervised tasks: Audio2Vec , which aims at reconstructing a spectrogram slice from past and future slices and TemporalGap , which
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::5d21974884cc8656f2edb3680563c065
https://doi.org/10.1109/lsp.2020.2985586
https://doi.org/10.1109/lsp.2020.2985586
Autor:
Ira Shavitt, Dotan Emanuel, Aren Jansen, Marco Tagliasacchi, Ronnie Maor, Joel Shor, Yinnon Haviv, Oran Lang, Felix de Chaumont Quitry, Omry Tuval
Publikováno v:
INTERSPEECH
The ultimate goal of transfer learning is to reduce labeled data requirements by exploiting a pre-existing embedding model trained for different datasets or tasks. The visual and language communities have established benchmarks to compare embeddings,
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0d18eac9b7f98a3f2f9cc45a39219640
http://arxiv.org/abs/2002.12764
http://arxiv.org/abs/2002.12764
Publikováno v:
SLT
This paper describes a new technique to automatically obtain large high-quality training speech corpora for acoustic modeling. Traditional approaches select utterances based on confidence thresholds and other heuristics. We propose instead to use an
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::66da1d4a5a69b299a352f6654f4662bf
https://doi.org/10.1109/slt.2016.7846323
https://doi.org/10.1109/slt.2016.7846323
Publikováno v:
ASRU
This paper describes a series of experiments to extend the application of Context-Dependent (CD) long short-term memory (LSTM) recurrent neural networks (RNNs) trained with Connectionist Temporal Classification (CTC) and sMBR loss. Our experiments, o
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::cf767228da74de6d095b4ab0947626a0
https://doi.org/10.1109/asru.2015.7404851
https://doi.org/10.1109/asru.2015.7404851