A neural machine translation (NMT) is an approach that uses an artificial neural network to determine matching and context. NMT systems progressively improve – and all parts of the system learn jointly – as they are exposed to more examples and content. They use a fraction of the memory that traditional statistical methods use. Google and Microsoft have adopted neural MT as their preferred methodology for machine translation. The Harvard NLP (Natural Language Processing) group has also developed an open source system called OpenNMT.

Receive exclusive access to members-only content, invitations to in-person and virtual events, and sneak peeks at new tools and emerging technologies. Join today!
Join Now!Categories
Authors
VIDEOS

Recent Posts
- How to fuel creativity in the workplace
- WHAT STORYTELLING MEANS TO VOICE AND TONE STRATEGY
- Viewpoint: Patient-focused Healthcare Information Requires an Omnichannel Content Strategy
- Viewpoint: The Disconnect Between Content Strategy, Information Science, and Healthcare
- The Need for Plain Language in “Terms and Conditions”