Training data is a resource used to develop machine learning models. In our definitive guide, we explain the best practices when creating your datasets and tips to improve your training data, as well as the best data annotation tools and open data resources.
Speech Data Collection for NICTLionbridge created a custom dataset of 300,000 non-native English speech samples for the National Institute of Information and Communications Technology. This data is being used to support the development of their translation app.Download
Text Classification for ZaizenLearn how Lionbridge provided high-quality text data for Zaizen, a company developing conversational AI systems that respond to user emotions. This text data was used to train a personalized AI for everyday conversations.Download
Text Classification for TravelokaLionbridge classified over 200,000 search queries for Traveloka, a leading online travel company. This data was used to build a search engine capable of returning results in multiple product categories.Download
Virtual Assistant DevelopmentFor an ongoing project with one of the world’s largest technology corporations, Lionbridge’s team edits and enhances the voice grammar framework for a leading virtual assistant.Download