Salesforce Just Open-Sourced a Large, XML-Tagged Machine Translation Dataset
Salesforce research on speeding up the work of post-editors by training MT engines with XML tags culminates in a 17-language online help dataset published on Github.
Salesforce research on speeding up the work of post-editors by training MT engines with XML tags culminates in a 17-language online help dataset published on Github.
Features 23 buyer profiles along industry verticals including buyer name, translation volume and / or spend (where available), technology used, sourcing approach, other key insights.
Teresa Marshall, Sr. Director of Localization and Globalization at Salesforce, explains how the CRM and cloud computing giant employs language strategies and partnerships for localization, and comments on their recent exploits in neural machine translation and its implications.