SwiftKey pioneered keyboards with a three-word suggestion bar above the keys that could accurately predict your next word. This was powered by a technology called “n-gram”, an approach now used on more than a billion devices globally.
N-gram technology has some limitations, as it can’t capture the underlying meaning of words and can only accurately predict words that have been seen before in the same word sequence. Now, SwiftKey Neural’s intelligent understanding of sentence context introduces a more ‘human’ touch for mobile typing.
Using machine learning and enormous amounts of language data, SwiftKey’s neural model is able to capture the meanings of words and the relationships between them. Within the neural model, words are organised in ‘clusters’, located at varying degrees of proximity to one another.
So, having seen the phrase “Let’s meet at the airport” during training, the technology is able to infer that “office” or “hotel” are similar words which could also be appropriate predictions in place of “airport”. Further, it understands that “Let’s meet at the airport” has a similar sentence structure to “Let’s chat at the office”. This intelligence allows SwiftKey Neural to offer the most appropriate next word based on the sentence being typed.
Until now, neural network language models have been deployed mostly on large servers, requiring significant computational resources. The launch of SwiftKey Neural Alpha is the first time this type of language model technology has been able to operate locally on a smartphone – a huge challenge given the resource constraints.
And the very clever people at SwiftKey reckon they are only scratching the surface of what’s possible with this technology.
