John Snow Labs, the Healthcare AI and NLP company and developer of the Spark NLP library, today announced that it will enable free access to its enterprise-grade Annotation Lab and NLP Server software for all users. This announcement comes on the first day of the company’s annual NLP Summit, a free online event that brings together the AI community to discuss the most important trends, use cases, and solutions advancing natural language processing (NLP).
The Annotation Lab, a robust data labeling and AI/ML solution for teams, enables users to annotate documents, images, and videos. The software automatically trains models using active learning and transfer learning. The simple and efficient project-based workflow helps users leverage real-time analytics on productivity, dataset bias, inter-annotator agreement, and more.
The enterprise solution also supports strong user management and role-based access, fully versioned and auditable completions, customizable workflows and tags, the ability to use custom models and embeddings, and integration with other tools using an API. Users can install the Annotation Lab within their own software infrastructure, including in air-gap networks, avoiding the need to share any data. There are no limits on the number of deployments, users, and customers.
This is significant, as many commercial offerings on the market are compliance- and cost-prohibitive, priced per document or user, and require data sharing. While open source solutions are a popular option, most charge extra fees for security, project management, and analytics features. With Annotation Lab, users have access to enterprise security, versioning, authentication, full audit trails, customizable workflows, and other critical features at no cost.
The NLP Server grants users access to thousands of pretrained out-of-the-box NLP models, enabling their use with no coding or data science experience necessary. Users can simply install the server as a single Docker container or on a cloud compute instance and access APIs. This eliminates apprehension around building pipelines or being versed in Python and other programming languages in order to leverage the models.
Valuable for both NLP novices and experts, the NLP Server allows users to send text and get results immediately or for data scientists to upload their own models. Teams can simply install and run the NLP Server within their own infrastructure, which prevents the need to share any data or models with a third party. Using the NLP server offers an easy solution to getting models and pipelines from the Spark NLP ecosystem to production quickly.
“Most NLP projects require not only the data science algorithms and models, but a way for domain experts to contribute their knowledge and quickly deploy trained models to production,” said David Talby, CTO, John Snow Labs. “By providing free enterprise-grade solutions to these problems — which were only accessible to companies with deep pockets until now — we hope to further democratize NLP and help the global AI community realize its potential faster.”
The Annotation Lab and NLP Server are constantly updated with new releases every two weeks to keep pushing state-of-the-art industry standards higher. John Snow Labs’ unique no data sharing design makes it a secure option for highly regulated industries, like healthcare, of which 60% of industry users rely on Spark NLP for their NLP projects.
To learn more about John Snow Labs’ recent contributions to the open-source community and the new free NLP software tools, join us at the NLP Summit. Free Registration is still open for the online conference being held from October 5-7.