Amazon Web Services Inc. said today it’s partnering with an artificial intelligence startup called Hugging Face Inc. as part of an effort to simplify and accelerate the adoption of natural language processing models.

Hugging Face is a New York City-based startup that’s best known for its Transformers library, which makes it easy to access a range of popular natural language neural networks trained on popular AI frameworks such as PyTorch and TensorFlow.

Transformers provides thousands of pre-trained models to perform tasks on texts, such as classification, information extraction, question answering, summarization, translation and text generation in more than 100 languages. Its aim is to make cutting-edge NLP easier to use for everyone.

Recognizing the popularity of Transformers, Amazon said today that it’s partnering with Hugging Face to combine that library with a number of its programming tools. The partnership will bring more than 7,000 NLP models to Amazon SageMaker, which is a machine learning service that’s used to build, train and deploy machine learning models.

For its part, Hugging Face has announced a couple of new services built using Amazon SageMaker, including AutoNLP, which provides an automatic way to train, evaluate and deploy state-of-the-art NLP models for different tasks, and the Accelerated Inference API, which is used to build, train and deploy machine learning models in the cloud and at the edge. The startup has also chosen AWS as its preferred cloud provider.

Amazon also announced the general availability of Hugging Face Deep Learning Containers, which is a service for developers to get started with building language models available through Hugging Face on Amazon SageMaker. The company said that with SageMaker, it’s possible to reduce the time it takes to perform NLP model experimentation from days to just minutes using an integrated development environment that helps to keep track of and compare the results of different experiments. In addition, developers can take advantage of SageMaker’s distributed training capabilities, Amazon said.

“Hugging Face is a resource for startups and other businesses around the world,” said Hugging Face Chief Executive Clement Delague. “Our transformers can help them build virtually any natural language processing application at a fraction of the time, cost and complexity they’d could achieve their own, helping organizations take their solutions to market quickly.”

Amazon said that some companies with early access have already used Hugging Face and SageMaker to create NLP models that can improve customer experiences. One such example is Quantum Health Inc., a healthcare navigation startup that uses AI to help consumers get access to better and more affordable health services.

“For some use cases we just use the Hugging Face models directly, and for others, we fine tune them on Amazon SageMaker,” said Jorge Grisman, an NLP data scientist at Quantum Health, who said the integration shortened the training time for its larger datasets.

Amazon said the net result of its partnership with Hugging Face is that customers will be able to train NLP models more easily, taking advantage of capabilities such as text generation, summarization, translation and conversation chatbots.

The collaboration with Hugging Face followed an earlier announcement from Amazon today about a new service that makes it easier to integrate AI into media content workflows. The newly launched AWS Media Intelligence Solutions is a combination of services that helps media creators to analyze their content, improve user engagement, reduce operational costs and increase the lifetime value of their content, the company said.

A blog post co-authored by Vasi Philomin, Amazon’s general manager of Machine Learning and AI, and Esther Lee, product manager for AWS Language AI Service, explained that AWS Media Intelligence Solutions incorporates a number of the company’s AI services. Amazon Rekognition, for example, is used to analyze images and videos, while Amazon Transcribe helps with audio transcription. Amazon Comprehend meanwhile provides natural language understanding, and Amazon Translate helps translate media content.

This combination of AI makes it possible to process and analyze media assets and automatically create metadata from images, video and audio content to be used in downstream workloads, Philomin and Lee said.

Amazon sees four primary use cases for AWS Media Intelligence: search and discovery, subtitling and localization, compliance and brand safety, and content monetization.

Image: Amazon