John Snow Labs Announces the Release of Spark NLP 2.7 with Hundreds of New Models, Capabilities
Jan. 13, 2021 — John Snow Labs, a Healthcare AI and NLP company and developer of the Spark NLP library, today announced that it has released Spark NLP 2.7, providing major enhancements to the open-source AI community. Known as the world’s most widely-used natural language processing (NLP) library in the enterprise, used by 54% of healthcare organizations, John Snow Labs attributes its growth and popularity to the consistent advancements made to its technology.
The release of Spark NLP 2.7 brings support to state-of-the-art Seq2Seq and Text2Text transformers. This includes new annotators for Google T5 (Text-To-Text Transfer Transformer) and MarianMNT for Neural Machine Translation, with over 646 new pre-trained models and pipelines. This provides users with fast, scalable, production-grade, open-source machine translation with just a few lines of code. Additionally, the new T5Transformer annotator and models achieve state-of-the-art results on multiple NLP tasks such as summarization, question answering, and sentence similarity.
This release also comes with new and improved models for language detection and identification that are faster, more accurate, and support 375 languages. Beyond adding and updating models for hundreds of languages, Spark NLP 2.7 introduces code changes and new algorithms for supporting a broader range of languages. For example, right-to-left languages such as Arabic, Farsi, Urdu, and Hebrew, are now fully supported. The new WordSegmenter annotator is trainable and provides out-of-the-box support for word segmentation in languages without rule-based tokenization, such as Chinese, Japanese, and Korean.
“Spark NLP 2.7 is the biggest release in John Snow Labs’ history, and one that we are thrilled to share with our global community,” said David Talby, CTO, John Snow Labs. “Democratizing state-of-the-art NLP through major open-source contributions and adding support for languages around the world to promote a more inclusive and diverse AI community is something we’re proud of and will continue to build on in the coming years.”
John Snow Labs’ commitment to providing access and availability of AI and NLP tools doesn’t stop at its technology offerings. The company also provides education and free resources to help further NLP knowledge and adoption, including the first Healthcare NLP Summit, being held from April 6-9. This event was created following John Snow Labs’ inaugural NLP Summit last year, which attracted more than 6,500 experts and practitioners in the field to discuss the trends, challenges, and best practices in NLP. The virtual event will be held again in October 2021.
To learn more about Spark NLP, or start your free trial, visit: https://www.johnsnowlabs.com/spark-nlp-health/.
Source: John Snow Labs