Microsoft has announced that it has reached the first technological breakthrough toward making conversations with chatbots similar to speaking with another person. The Redmond-based software giant was one of the first to release a chatbot to directly interface with normal people. However, it did not go down well and the chatbot was soon found to be sending abusive tweets and making racial slur. Microsoft soon pulled the plug on the chatbot and released it few days later with a fix.
Since that incident, Microsoft seems to have come a long way with its chatbot development. Microsoft today announced that it has figured out how to make chatbots talk and listen at the same time. It says the new enhancement will allow chatbots to operate in full duplex mode, borrowing a term from telecommunications. The software giant says the feature will allow chatbots or any digital assistant to have a flowing conversation with humans and reduce the feeling of having a conversation with a bot.
The chatbots and digital assistants currently work in a way where only one side can talk and most of the replies are programmed and not continuous or instantaneous. Microsoft says the technology is already up and running in Xiaoice, Microsoft’s AI chatbot currently available in China. Microsoft says Xiaoice can quickly predict what the person it is speaking to will say using the “full duplex voice sense.”
“That helps her make decisions about both how and when to respond to someone who is chatting with her, a skill set that is very natural to people but not yet common in chatbots,” Microsoft says. “This really speeds up her responses to be much more natural,” said Ying Wang, a Microsoft director who oversees Zo.
Watch: Xiaomi Redmi 5 First Look
The new breakthrough also gets rid of using a “wake word” needed to interact with chatbots. The company plans to bring the same technology next to its chatbot Zo in the US before expanding it to other social chatbots. Microsoft has already achieved human-level parity on speech recognition and translation and is increasing focusing on AI and machine learning. The company also reorganized to bring AI and cloud to the forefront last week.