Researchers at Microsoft claimed to
have developed the first technology that recognizes the words in a conversation
as well as humans do. A team in Microsoft Artificial Intelligence and Research
created a speech recognition system that makes the same or fewer errors than
professional transcriptionists. The system had a word error rate (WER) of 5.9%
- the lowest ever recorded against the industry standard Switchboard speech
recognition task. The research milestone does not mean the computer recognized
every word perfectly. In fact, humans do not do that, either. Instead, it means
that the error rate – or the rate at which the computer misheard a word like “have”
for “is” or “a” for “the” – is the same as you would expect from a person
hearing the same conversation. The milestone means that, for the first time, a
computer can recognize the words in a conversation as well as a person would. The
milestone comes after decades of research in speech recognition, beginning in
the early 1970s with DARPA, the US agency tasked with making technology
breakthroughs. This accomplishment is the culmination of over twenty years of
effort. The milestone will have broad implications for consumer and business
products that can be significantly augmented by speech recognition. That includes
entertainment devices like the Xbox, accessibility tools such as instant speech
to text transcription and personal digital assistants such as Cortana.
No comments:
Post a Comment