Towards learning a universal non-semantic representation of speech

J Shor, A Jansen, R Maor, O Lang, O Tuval… - arXiv preprint arXiv …, 2020 - arxiv.org
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 …
被引用次数：154 相关文章

Towards Learning a Universal Non-Semantic Representation of Speech

J Shor, A Jansen, O Lang, O Tuval… - Proc …, 2020 - research.google
The ultimate goal of transfer learning is to enable learning with a small amount of data, by
using a strong embedding. While significant progress has been made in the visual and …

Towards Learning a Universal Non-Semantic Representation of Speech

J Shor, A Jansen, R Maor, O Lang, O Tuval… - arXiv e …, 2020 - ui.adsabs.harvard.edu
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 …
[PDF] interspeech2020.org

[PDF][PDF] Towards Learning a Universal Non-Semantic Representation of Speech

J Shor, A Jansen, R Maor, O Lang, O Tuval… - 2020 - interspeech2020.org
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 …
[PDF] isca-archive.org

[PDF][PDF] Towards Learning a Universal Non-Semantic Representation of Speech

J Shor, A Jansen, R Maor, O Lang, O Tuval… - 2020 - isca-archive.org
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 …

Towards Learning a Universal Non-Semantic Representation of Speech

J Shor, A Jansen, O Lang, O Tuval… - Proc …, 2020 - research.google
The ultimate goal of transfer learning is to enable learning with a small amount of data, by
using a strong embedding. While significant progress has been made in the visual and …