Unbabel’s machine translation technology might be the best any startup has to offer right now
Portuguese startup Unbabel, which we profiled here earlier this year, has raised a fresh $5 million in a Series A round led by Notion Capital and Portuguese PE/VC firm Caixa Capital, the company announced Monday. The round also included returning investors Shilling Capital Partners and Faber Ventures.
The cash will go toward further building out its natural language processing (NLP) platform, the Unbabel Language Engine. The engine combines the work of some 40,000 human translators and editors with a combination of artificial intelligence and machine translation mechanisms. They claim to have brought the cost of translation down by a factor of ten.
“With a $38 billion market in translation services, it’s clear that the expense is huge for any company looking to do business in a multilingual world,” Unbabel Co-Founder and CEO Vasco Pedro asserted in a press release. “However, this market is comprised of an incredibly long tail of high-cost, low-tech, and massively inefficient solutions where the business model is basically ‘throw more humans at it.'”
“We still think humans play a hugely important role in the process, but much further up the stack, atop cutting edge tech that does 95% of the job for them. With this approach, we can translate text faster and cheaper, including types of content that were previously seen as untranslatable like emails and chat,” he noted.
Unbabel is playing with some big-name clients at this point, including Microsoft (which has its own MT tool, Microsoft Translator), Zendesk, Salesforce, Pinterest, and Skyscanner. They offer APIs to combine with companies’ own content (and in-house translation) workflows.
“Machine learning sucks and the translation is not there. Human translation is expensive and not scaleable,” CEO Vasco Pedro told Geektime earlier this year. “[Our tech] very thoroughly automates the process of language but uses humans in the parts you need to use, like the creative parts. Machine translation is the initial basis.”
The company, which was founded in 2013 and has raised $8 million to date including this current round, is one of several strong startups worthy of its reputation in machine translation (MT). This industry is expected to grow at 23.53% annually through 2019 according to Research and Markets, perhaps to as much as $983.3 million by 2022 according to Grand View Research. The most widely used tool on the market right now is free, Google Translate, but it can’t compete with the likes of Unbabel — yet.
“We have a fundamentally different approach from Google,” Vasco relayed to Geektime in February. “They’re doing an amazing job creating a general purpose system. We started a general system but then adapt it to particular domain and [a] particular customer that resolves a lot of the ambiguities in general systems.”