Wluper, the London-based tech startup building a conversational AI to power knowledge-based voice assistants, has raised $1.3 million in seed funding. Leading the round is “deep tech” VC IQ Capital, with participation from Seedcamp, Aster, and Magic Pony co-founder Dr Zehan Wang.
Founded in 2016 and originally backed by Jaguar Land Rover’s InMotion Ventures, Wluper’s “conversational AI” is initially targeting navigation products with what it describes as “goal-driven dialogue” technology that is designed to have more natural conversations to help with various navigation tasks.
The ‘secret sauce’, as it were, is that Wluper believes voice assistants work much better when the underlying AI is tasked with becoming an expert in a more narrow and specialist domain.
“When we think of intelligent assistants like Alexa or Siri, the only time you’ll believe they’re really good is if they understand you properly; most of the time, they simply can’t,” says Wluper co-founder Hami Bahraynian. “It is not the speech recognition which fails. It is the missing focus and lacking reasoning of these systems, because they all can do a lot of things reasonably well, but nothing perfectly”.
Describing the goal of “general” conversation AI as one that could take 15, 20 or more years to achieve, Bahraynian says that in the interim what is needed is “intelligent agents” that are created for a certain purpose, now.
“This is exactly what we do,” he says. “We build domain-expert conversational intelligence, which does one thing, understanding everything transport-related, but that one thing perfectly”.
Furthermore, Wluper’s approach is able to make clear assumptions regarding what the user is talking about, and therefore claims to be able to understand much more complex questions and in a more natural way. This includes multi-intent queries, and follow-up questions to enable a “true” conversation, says Bahraynian.
In addition, Wluper has been conducting R&D in what comes after the “understanding” bit of the NLP pipeline, leading the startup to undergo further research on a machine’s “knowledge acquisition” capabilities, which it believes is a crucial piece of the puzzle needed to solve conversational AI.
“Even if naturally asked user queries are eventually understood correctly, extracting and providing relevant and useful information from the right places is even more challenging, and with current mostly ruled-based approaches, ultimately impossible to scale,” adds Bahraynian.
“We work on this problem by moving away from traditional handcrafted methods and work on new ways to optimise a machine’s knowledge acquisition and finding the right balance between structured and unstructured data in order to provide more meaningful results”.
Meanwhile, Wluper’s seed investment will be used to hire more engineers and research scientists to expand the startup’s research and development capabilities.