ELSA, an application whose name means “English Language Speech Assistant” (and not the famous Disney character!), Raised $ 3.2 million for its supported language learning platform by computer. Unlike other courses that focus primarily on teaching grammar and vocabulary, ELSA uses artificial intelligence and speech recognition technology to help language learners to pronounce their language skills. pronunciation.
The $ 3.2 million funding granted before the end of the funding cycle was led by Monk’s Hill Ventures, a company that invests in start-up companies in Southeast Asia. The founder and associate of Monk’s Hill, Peng T. Ong, joins the ELSA board.
The San Francisco-based start-up was founded in 2015 by Vu Van, CEO of ELSA, CEO of ELSA, and Dr. Xavier Anguera, whose background is in voice recognition and A.I. technologies. He made his debut at SXSW in March 2016, where he later won the SXSWedu Launch Competition.
According to Van, who was born and raised in Vietnam, the idea of ELSA was inspired by her personal experience in trying to learn English.
“I moved to the United States for my MBA and Master’s degree in Education at Stanford,” she says. “My first year at Stanford was very difficult because of my ability to speak English, and most of the time people misunderstood me,” says Van.
Although the founder of ELSA was able to write and read English quite well, she wanted to find a good solution to improve her accent – and she came back to reality.
“I realized that people did not really have a lot of solutions … when it was talking, they could talk to a speech therapist who was costing them 150 $ time, who could listen to them and correct their pronunciation, or they could go on YouTube or watch Netflix, which is a one-way learning solution, “says Van.
She decided to develop ELSA as a result of her own struggles, bringing co-founder Dr. Anguera to help create ELSA’s proprietary speech recognition technology.
To use ELSA, language learners download the app for iOS or Android, and then take the ELSA five-minute assessment test that identifies the pronunciation skill of the speaker. user and identifies the areas where they still have difficulties. This information is then used to develop a personalized program, tailored to the current capabilities of the user.
In ELSA, there are about 600 lessons and more than 3,000 words on various topics, such as introductions, family or relationship discussions, jobs and hiring, travel and more . The app is also updated regularly with seasonal and timely topics, like lessons on the holidays, or even the new movie “Star Wars”, so that learners can better participate in daily communication.
The lessons themselves are small, 2 minutes long. They have five exercises that are getting more and more difficult, including pronunciations of words, sentences and sentences.
ELSA works by listening to the learners’ voices, and then matching what they said with the correct American English pronunciation. The words on the screen are highlighted in red, yellow and green to show how well each student has done. The application will also help by making suggestions as to how the speaker can improve a given sound – for example, by telling them how to shape their mouth or move their tongue.
Behind the scenes, ELSA is powered by A.I. technology, which listens to users’ speech – something that the team has built from scratch.
“Speech recognition technology is slightly different [from ELSA] They try to guess what you say, whether you say it right or wrong, it’s obviously very forgiving to make mistakes, what we want to do, it’s exactly the opposite, “Van says.
Since its launch two years ago, ELSA has been adopted by a few million users in more than 100 countries, with about half of its user base in Southeast Asia and the United States. stays elsewhere in the world. Europe . Users nowadays practice a few million exercises a week.
With the additional funding, ELSA plans to meet the demands of teachers who want to use the service in their classrooms by developing a tool that will allow them to enter words and phrases from their own program. He will also begin work to add new languages to the platform and build his speech recognition API.
ELSA now generates revenue through subscriptions, priced at $ 3.99 / month or $ 29.99 / year. A quarterly option is also available, but prices may vary by region. Although the product for teachers is affordable, it will not be free – potentially offering another revenue stream.
The company will also use the funds to hire other talent A.I., says Vu.
The ELSA app is a free download on iOS and Android.