While the familiar idiom “you say tomayto, I say tomahto” is meant to showcase the triviality of differences, the irony lies in its illustration of the wide variation in English pronunciation.

Such vagaries in pronunciation can make English difficult for many nonnative speakers unused to pronouncing certain sounds. English is a stress-based language, meaning that it requires emphasis on particular syllables, said Sarah Daniels, CEO and co-founder of English-learning startup Blue Canoe. “If someone is not proactively thinking about stress … we, in our system, can teach them where it is and how to do it.”

Bellevue-based Blue Canoe’s mobile app directs its users to repeat sentence prompts and record them. Speech-recognition technology then analyzes the recordings and uses machine-learning models to point out the differences. When users spend 10 minutes per day on the app, personalized feedback from an artificially intelligent (AI) teacher informs students precisely how they mispronounced words.

The A.I. Age | This 12-month series of stories explores the social and economic questions arising from the fast-spreading uses of artificial intelligence. The series is funded with the help of the Harvard-MIT Ethics and Governance of AI Initiative. Seattle Times editors and reporters operate independently of our funders and maintain editorial control over the coverage.

The startup plans to expand its services in East Asia with the help of $2.5 million in seed funding raised earlier this month. Since last year, Blue Canoe has sold its services to multinational companies in the professional services and tech sectors that provide the app to their employees. English-teaching companies in Japan, China and the U.S. also offer the app to clients for additional training. The round of seed funding led by Tsingyuan Ventures, Qualcomm Ventures and others brings the total raised by the company to $3.9 million.

To help users learn where to place the correct stress on syllables in real time, the startup digitized a 20-year-old brain-based methodology called the Color Vowel System, then hired linguists to listen to users’ recordings and tag the problems. The amount of tagged dataset recordings increases every time the app is used, which helps the machine-learning models continuously improve.

Research has shown that simply reciting words doesn’t help English learners’ pronunciation, because their brains have already been programmed to understand their native tongue, said Daniels. “With (the app’s) targeted feedback, it is head-and-shoulders more effective for somebody to cognitively understand what they did wrong and make the corrections in order to improve.”


While Daniels noted there is no “right” accent for English, she said the sounds and rhythms of a person’s native language may make it difficult to understand their speech.

That was the case for a 15-year-old native Mandarin speaker at the English tutoring company We Education, who repeated words his tutor said in English with little improvement in his pronunciation for a year. The words were brought to life for the student based in Anhui, China, when We Education began testing and piloting Blue Canoe services in May, said the company’s CEO, George Cigale. “Once the tutor introduced the Color Vowel Method, the student could physically feel where the stress is and visually see where the tongue should be placed to pronounce the correct vowel sound,” Cigale wrote in an email.

The Tokyo office of multinational accounting firm Ernst & Young tried several English-language programs that were ineffective in improving their employees’ spoken English. Since purchasing Blue Canoe services last December, Ernst & Young spokesperson Neil Hasegawa-Yates said the real-time feedback that’s given to employees has instilled in them confidence in their stress and timing of spoken English.

“The switch from trying to speak English with a Japanese toolbox, to suddenly realizing there is a whole new toolset available to play with, takes a lot of the frustration out of trying to master a second language,” he said.

The recent round of seed funding came after the startup won $500,000 at Qualcomm’s inaugural Female Founders Summit pitch competition held in July in San Francisco. Quinn Li, senior vice president and global head of Qualcomm Ventures, said Blue Canoe’s traction with customers in Fortune 500 companies and Daniels’ entrepreneurial skills made the startup an easy pick. Blue Canoe is the fifth startup Daniels has been involved in in the past two decades.

Li hopes Qualcomm’s partnerships with multinational companies and English-tutoring schools will help scale the startup. “We have a broad workforce in countries outside the U.S., and these employees could certainly benefit from their product.”

In the meantime, Daniels said she regularly receives emails from users that make her day. One customer shared the tribulations of being misunderstood after living in the U.S. for over a decade. After using Blue Canoe, the user was able to get on a bus one day and to be easily understood by the driver without the usual back-and-forth. “I have a lot of empathy for people who are learning new languages and trying to get their jobs done, get on a bus or buy something in a store,” said Daniels. “When we can help with that, it can make a pretty big difference in their lives.”

Correction: This story has been updated to clarify Sarah Daniels’ involvement in start ups.