Research shows that robots with local accents appear more reliable and competent to users

Whether it’s Optimus Prime or the Daleks from Doctor Who, most robots have the same monotone, automated voice.

But research shows that certain groups of people prefer it when they use a familiar accent or dialect.

Research has shown that speaking with a local accent can, under certain circumstances, make robots appear more reliable and competent.

Scientists from the University of Potsdam, in Germany, recruited 120 people from Berlin or Brandenburg to complete an online survey.

They asked participants to watch videos in which a robot spoke with a male human voice in standard German or Berlin dialect, which is considered working class.

Most robots – like the Daleks and Optimus Prime – have the same monotone, automated voice

Research has found that speaking with a local accent – ​​such as the Berlin dialect (pictured) – can make robots seem more reliable and competent

Research has found that speaking with a local accent – ​​such as the Berlin dialect (pictured) – can make robots seem more reliable and competent

People were asked to rate the robot’s reliability and competence, and to complete a questionnaire including items about how well they spoke the Berlin dialect and how often they used it.

In general, respondents preferred a robot that speaks standard German.

But those who felt more comfortable with the Berlin dialect preferred the robot to speak their local language.

Lead author Katharina Kühne said: ‘If you are good at speaking a dialect, you are more likely to trust a robot that talks the same way.

‘It seems that people trust the robot more because they find a similarity.’

The team said context can play an important role in determining when a local accent is appropriate.

For example, it could be useful in a nursing home scenario where it is important that people feel connected to the robot.

But in other situations – such as in a hospital or hotel – people may prefer a robot that speaks with a standard accent.

The findings were published in the journal Frontiers in Robotics and AI.