3 ex-Stanford students are working on an app to change your accent

By Catherine E. Schwecht | CNN

The Stanford University students heard grief in their friend’s voice as he shared the news.

“Guys, I had to quit my job.”

For them, it just didn’t make sense. He was fluent in both English and Spanish, very friendly and an expert in systems engineering. Why wasn’t he able to hold a job at the call center?

The friend said his accent made it difficult for many clients to understand him; Some even threw insults because of the way he spoke.

The three students realized that the problem was bigger than their friends’ experience. So they set up a startup to solve it.

Now, their company, Sanas, is testing AI-powered software that aims to eliminate miscommunication by changing people’s accents in real time. A call center worker in the Philippines, for example, can normally speak into the microphone and end up looking like someone from KS to a customer on the other end.

The startup’s founders say call centers are only the beginning. The company’s website promotes its plans as “talk reimagined.”

Ultimately, they hope that the app they are developing will be used by a variety of industries and individuals. It might help doctors better understand patients, they say, or help grandchildren better understand their grandparents.

“We have a very big vision for Sanas,” CEO Maxim Serebryakov says.

And for Serebryakov and his co-founders, the project is personal.

This screenshot shows what users see in the Sanas desktop app (Image courtesy of Andrés Pérez Soderi via CNN)

You don’t hear people’s voices as much as you hear their accents

The trio who founded Sanas met at Stanford University, but they are all from different countries originally — Serebryakov, now CEO, from Russia; Andrés Pérez Sodiri, now Chief Financial Officer, is from Venezuela. and Shun Zhang, now chief technology officer, from China.

They are no longer Stanford students. Serebryakov and Peres graduated; Chang pulled out to focus on bringing Sanas back to life.

They launched the company last year, giving it a name that can be easily pronounced in different languages ​​”to highlight our global mission and desire to bring people together,” says Perez.

Over the years, all three say they have experimented with how dialects can get in their way.

“We all come from international backgrounds. We’ve seen firsthand how people treat you differently just because of the way you speak,” says Serebryakov. “It’s heartbreaking sometimes.”

Zhang says his mother, who came to the US more than 20 years ago from China, still makes him talk to the cashier when they go grocery shopping together because she feels embarrassed.

“That’s one of the reasons I joined Max and Andrés in building this company, and trying to help those people who think their voices aren’t heard as much as their accents,” he says.

Serebryakov says he has seen how his parents are treated in hotels when they come to visit him in the US – how people make assumptions when they hear their accents.

“They talk a little louder. They change their behaviour.”

Perez says that after attending a British school, he initially struggled to understand American dialects when he arrived in the United States.

And don’t get it started on what happens when his dad tries to use the Amazon Alexa that his family gave him for Christmas.

“We quickly found out, when Alexa was turning on lights in random places in the house and making them pink, Alexa didn’t understand my dad’s accent at all,” Perez says.

Call centers are testing technology

English is the most spoken language in the world. It is estimated that 1.5 billion people speak it – and most of them are not native speakers. In the United States alone, millions of people speak English as a second language.

This has created a growing market for apps that help users practice their English pronunciation. But Sanas is using AI to take a different approach.

The premise: Instead of learning to pronounce words differently, technology can do it for you. There will no longer be a need for costly or time-consuming training to reduce accent. And the comprehension will be almost instant.

Serebryakov says he knows people’s dialects and identities can be closely linked, and stresses that the company isn’t trying to erase dialects, or suggest that one way of speaking is better than another.

“We allow people to not have to change the way they speak to occupy a position, to hold a job. Identity and dialects are critical. They are intertwined,” he says. “You never want someone to change their tone just to please someone.”

Sanas’ algorithm can currently convert English to and from American, Australian, British, Filipino, Hindi and Spanish dialects, and the team plans to add more. They can add a new tone to the system by training a neural network with audio recordings from professional actors and other data – a process that takes several weeks.

The Sanas team played two CNN shows. In one, a man with an Indian accent was heard reading a series of literary sentences. These same phrases are then converted into American dialect:

Another example with phrases that might be more common in a call center setup, such as “If you give me your full name and order number, we can go ahead and start making the correction for you.”

The results with an American accent sound somewhat artificial and integrated, like virtual assistant voices like Siri and Alexa, but Perez says the team is working to improve the technology.

“The tone changes, but the intonation is preserved,” he says. “We are continuing to work on how to make the result as natural, emotional and exciting as possible.”

Perez says early feedback from call centers that have tried this technology has been positive. So submit comments on their website as word spreads about their project.

They say their plans for the company generated $5.5 million in seed funding from investors earlier this year.

How do the founders of the startup see its future

This allowed Sanas to increase the number of its employees. Most of the employees of Palo Alto, based in California, come from international backgrounds. This is not a coincidence, says Serebryakov.

“What we build has resonated with a lot of people, even the people we hire. It’s really exciting to see,” he says.

As the company grows, it may take some time before Sanas appears in the App Store or appears on a mobile phone near you.

The team says they are working with the largest call center outsourcing companies at the moment and are choosing slower rollouts for individual users so they can improve technology and ensure security.

Leave a Comment