that Munster split the questions into two halves, asking 800 on a busy street and 800 in a quiet room. Munster then published his results in a note to clients on Thursday.

When testing Google, the search engine understood 100 per cent of the typed in questions and replied accurately 86 per cent of the time, earning a B+ from the analyst.

understood 83 per cent of the queries in noisy conditions, and 89 per cent in a quiet room. Siri gave an accurate reply 62 per cent of the time on the busy streets and 68 per cent in a quiet environment. This earned Siri a grade D for accuracy, according to Munster.

"In order to become a viable mobile search alternative, Siri must match or surpass Google's accuracy of B+ and move from a grade D to a B or higher," wrote Munster.

The release of iOS 6 this autumn and the decrease in reliance on Google could help Siri catch up, said Munster, who believes that the iPhone's assistant is two years behind Google in its .