Apple’s voice recognition platform appears to hold muster.
But the praise probably doesn’t deserve to go much higher than that. According to a new study, the accuracy level of Siri was placed at 62%.
Piper Jaffray analyst Gene Munster took a team of testers and in aggregate they placed 800 queries to Siri.
Questions were posed in both a quiet room and out on the streets to see if ambient noise had any impact on the success of the service.
According to Munster, in the outdoor environment, Siri was able to comprehend the questions around 83% of the time, but its accuracy in providing answers was only 62%.
As for the examples in a quiet room, it performed slightly better, with comprehension at 89% and accuracy at 68%.
By means of example, when asked, “Where is Elvis buried?” Siri provided users with details of people named “Elvis Buried” and when asked, “When did the movie Cinderella come out?” Siri attempted to find current movie times.
To compare, Munster and his team also posed the same questions to Google. Of course, in these instances the questions were entered via text, but the accuracy level was much higher – 86%.
“We believe the most likely substitute for Siri is Google text input on the iPhone, not the Google voice search app,” said Munster.
As for Siri, though, Munster had good things to say. “While Siri is two-plus years behind Google today, we are optimistic about its future. With the iOS 6 release in the fall, we expect Siri to improve meaningfully while reducing its reliance on Google,” he said.
