Researchers at software giant Microsoft claim to have created what the world's first machine translation system capable of interpreting and translating sentences from Chinese to English with the same quality and accuracy as a person.
The researchers, based in Microsoft's Asia and US labs, said in a blog post that their system achieved human parity on a commonly used test set of news stories, called Newstest2017, a system developed by a group of academics.
To ensure the results were both accurate and on a par with what people would have translated, the team hired external bilingual human evaluators, who compared Microsoft's results to two independently produced human reference translations.
Hitting human parity in a machine translation task is a dream that all of us have had... We just didn't realise we'd be able to hit it so soon
The technical fellow in charge of Microsoft's speech, natural language and machine translation efforts, Xuedong Huang, described it as a "major milestone" in one of the most challenging natural language processing tasks.
He continued: "Hitting human parity in a machine translation task is a dream that all of us have had," Huang said. "We just didn't realise we'd be able to hit it so soon."
Huang, who also led the group that recently achieved human parity in a conversational speech recognition task, said the translation milestone was especially unique because of the possibilities it has for helping people understand each other better.
"The pursuit of removing language barriers to help people communicate better is fantastic," he said. "It's very, very rewarding."
Machine translation is a challenge researchers have worked on for decades. And, experts say, for much of that time many believed human parity could never be achieved. Still, the researchers cautioned that the milestone does not necessarily mean that machine translation is a solved problem.
Ming Zhou, assistant managing director of Microsoft Research Asia and head of a natural language processing group, said that while the team was thrilled to have achieved the milestone on the dataset, he added that there are still a number of challenges ahead, such as testing the system on real-time news stories.
Allen died from complications of non-Hodgkin's lymphoma
Stanford researchers made the discovery via data from Greenland
Created via a thin, flexible, and transparent hierarchical nanocomposite film
Rolls Royce will use AI powered by Intel's Xeon Gold processors and SSDs for memory