Scots researchers modify androids voices to make them sound more trustworthy

But those that have a more ‘sing song’ one with personality and intonation gave the listener a good impression of the speaker.

The findings opens the door for online retailers such as Google, Amazon and Apple to sell us more things by humanising the robotic voices increasingly being used in their businesses.

It could also help young humans learn how to come across well in interviews as three fifths are now carried out over the phone.

Previous research has shown a person’s voice can give out clues to their authoritiveness, career success, fertility, height and attractiveness - in a split second.

It has been rumoured Margaret Thatcher and the Queen were both trained to make their voices appear more dominant.

Psychologist Dr Phil McAleer said people often get just one chance to make a first impression - and this is even harder when it is just their voice they are judged on.

Now his team at the University of Glasgow have found a way to modulate a voice pattern which a listener perceive as “trustworthy” which they hopes will lead to a new synthetic voice personality that would give the best first impression to listeners.

It could also have applications from everything from your car’s SatNav to automated telephone banking.

The ‘Sound of Trustworthiness’ described in the journal PLOS used computer modelling and web-based judgments to generate voices and gauge how honest they seemed to listeners.

Dr McAleer said: “We were really surprised by how faithfully the personality impressions followed our acoustical manipulation. It was almost like we had a sliding scale of trustworthiness.

“The more monotone or flat the voice sounded, the less appealing it was to the listener. While a voice with a bit more personality and intonation gave the best and most positive impression.

“It was as if we were turning a ‘voice trustworthiness’ button on a sound system.”

Research shows when we hear a new voice we form personality impressions, such as trustworthiness or dominance.

These “first impressions” are quite consistent across listeners, even in a single word like “hello”.

But the acoustical basis for this phenomenon has remained elusive, such as how exactly should one say “hello” to be perceived as trustworthy by new listeners.

The scientists have now solved that problem.

Lead author Professor Pascal Belin, of Aix-Marseille University, France, said: “These results demonstrate there is a strong acoustical basis to our voice personality impression. We have now a principled method to modulate the personality impressions that voices elicit.

“Our results were obtained with trustworthiness but there is no reason why it should not work with other personality traits.”

In the study the researchers first generated two synthetic voice prototypes by averaging recordings from individuals rated high or low in perceived trustworthiness.

Then they generated a continuous series of voices between these two prototypes which they asked 500 volunteers to rate on trustworthiness via a web interface.

Dr McAleer says a man who raises his pitch becomes more trustworthy whereas it works the opposite way for a woman who is seen as less trustworthy if she glides up at the end of a word.

Now, he said it could help tech giants such as Google, Amazon and Apple to sell things as they are increasingly incorporating robotic voices into their business.

He said: “If they want people to be interactive with them they will need to sound like you can trust them. Also, it could help younger people train for interviews. They are told about their clothes and their hair all the time, but the voice is always overlooked.

“And 60 per cent of interviews are now carried out over the phone. It maybe unfair if they are judged on their voice, but they need to be aware of it.”

BACKGROUND: Scots hit the perfect pitch as their accents give the ring of honesty

By Sandra Dick

KEVIN Bridges tells the story of standing in a bar somewhere in Spain.

“Yous still sellin’ fid?,” he asks in his best Glaswegian. “Que?” responds the confused waiter.

The Englishman standing next to him gleefully translates. “He asked you. Are. You. Still. Serving. Fooood?”

Bridges sums up. “I’ve gone on holidays where I’ve had people translate for me … into English.

“Everywhere you go as a tourist, people speak English. When you’ve got a Scottish accent, well, that’s very little help.”

That may be the case, but it would appear Scots, with our typically gentle accents and singsong lilt, could actually have the upper hand when it comes to connecting with consumers.

Over recent decades Scotland has emerged as a call centre capital of the world, with more than 100,000 employees working in at least 400 contact centres, from banking and financial services to utilities and technical support.

Across the central belt, one in every 30 workers is employed in some kind of call centre capacity.

Indeed, virtually every business sector has a Scot somewhere on the end of the phone.

It would appear that while we may not be too good at reaching the finals of the World Cup, we are premier league material when it comes to talking.

Researchers at the University of Glasgow have now discovered that modifying artificial computer voices to give them a “sing song” voice appears to be the key to making consumers respond best.

While the light lilting tones create an air of trustworthiness, a dull monotone on the other end of the line is likely to be met with a swift “click … burr …”

Pinpointing this elusive “sound of trustworthiness” could spell big business.

From Siri to driverless cars, the way we interact with inanimate objects is developing at a rapid pace – and the trust we place in them is fundamental to their success.

Researchers at the University of Chicago Booth School of Business recently put “trust” to the test amid growing concerns from the autonomous car developers that drivers simply do not trust vehicles that drove themselves. They inserted a human voice into one vehicle and gave it a gender and a name, Iris. They found people reacted differently when dealing with Iris than a nameless, voiceless car, even down to defending Iris’s when the vehicle was involved in a collision.

Typically, the vehicle without a voice and a nice name, received little sympathy.

But of course some may well argue that what’s said is more important than how it’s said.

Psychologist Dr Cynthia McVey points out that voices are a combination of tone, accent and pitch.

“The gentler the tone and pitch, the easier they are to listen to,” she adds.

“Then some of what’s seen as trustworthy might come down to familiarity, and some to characteristics.”

She added: “Scots are often seen as careful with their money, perhaps more honest, more straight talking. “Hearing that accent may make people feel it’s someone they can trust.”