Magenta
| ISIN | 🆔 |
|---|---|
| Industry | Artificial intelligence |
| Founded 📆 | 2017 |
| Founder 👔 | |
Area served 🗺️ | |
| Owner | |
| Members | |
Number of employees | |
| 🌐 Website | magenta.tensorflow.org |
| 📇 Address | |
| 📞 telephone | |
Magenta is a public Google research team exploring the creative applications of machine learning and artificial intelligence.[1] The group was started by researchers and engineers from the Google Brain team. [2]
Research[edit]
Magenta is researching how machine learning algorithms, data sets and models can be used as a tool in the creative process by artists and musicians.[3] Their research papers are publicly available from the Google Research website [4] and other open-access repositories such as arXiv [5][6] The majority of those papers investigate the role of machine learning in different aspects of music, such as composition, synthesis and generation.[7]
Datasets and models[edit]
As part of their research, Magenta also released a series of publicly available data sets and models[8], including:
- Nsynth - An audio dataset including more than 305,979 musical notes, for 1,006 instruments. Notably used by the open source project NSYNTH SUPER.[11][12]
- Quick, Draw! - A collection of over 50 millions of drawings, submitted by users while playing the Quick, Draw! online game.[13]
References[edit]
- ↑ "Make Music and Art Using Machine Learning". Archived from the original on 2020-02-19. Retrieved 2020-02-29. Unknown parameter
|url-status=ignored (help) - ↑ "Magenta". Retrieved 2020-02-29.
- ↑ "Make Music and Art Using Machine Learning". Archived from the original on 2020-02-19. Retrieved 2020-02-24. Unknown parameter
|url-status=ignored (help) - ↑ "Publication database". Retrieved 2020-02-24.
- ↑ Choi, Kristy; Hawthorne, Curtis; Simon, Ian; Dinculescu, Monica; Engel, Jesse (2019). "Encoding Musical Style with Transformer Autoencoders". arXiv:1912.05537 [cs.SD].
- ↑ Cheng-Zhi Anna Huang; Vaswani, Ashish; Uszkoreit, Jakob; Shazeer, Noam; Simon, Ian; Hawthorne, Curtis; Dai, Andrew M.; Hoffman, Matthew D.; Dinculescu, Monica; Eck, Douglas (2018). "Music Transformer". arXiv:1809.04281 [cs.LG].
- ↑ "Datasets". Archived from the original on 2019-08-27. Retrieved 2020-03-18. Unknown parameter
|url-status=ignored (help) - ↑ "Datasets". Archived from the original on 2019-08-27. Retrieved 2020-02-24. Unknown parameter
|url-status=ignored (help) - ↑ "The MAESTRO DatasetContentsDatasetDownloadLicenseHow to Cite". Archived from the original on 2019-07-14. Retrieved 2020-02-24. Unknown parameter
|url-status=ignored (help) - ↑ Hawthorne, Curtis; Stasyuk, Andriy; Roberts, Adam; Simon, Ian; Cheng-Zhi Anna Huang; Dieleman, Sander; Elsen, Erich; Engel, Jesse; Eck, Douglas (2018). "Enabling Factorized Piano Music Modeling and Generation with the MAESTRO Dataset". arXiv:1810.12247 [cs.SD].
- ↑ "The NSynth DatasetDownloadContentsMotivationDescriptionFormatStatisticsLicenseHow to CiteUpdates". Archived from the original on 2019-09-02. Retrieved 2020-02-24. Unknown parameter
|url-status=ignored (help) - ↑ "Making music using new sounds generated with machine learning". Archived from the original on 2020-02-17. Retrieved 2020-02-24. Unknown parameter
|url-status=ignored (help) - ↑ "What do 50 million drawings look like?". Archived from the original on 2020-01-14. Retrieved 2020-02-24. Unknown parameter
|url-status=ignored (help)
External links[edit]
 | This Google-related article is a stub. You can help EverybodyWiki by expanding it. |
This article "Magenta (Google Research)" is from Wikipedia. The list of its authors can be seen in its historical and/or the page Edithistory:Magenta (Google Research). Articles copied from Draft Namespace on Wikipedia could be seen on the Draft Namespace of Wikipedia and not main one.
