Users can train systems on “multi-speaker data using personality embeddings”
Microsoft has open sourced its latest contribution to the conversational AI sector: a toolkit to help developers “imbue their chatbots with different personas”.
The release, dubbed Microsoft Icecaps, comes as developments in natural language processing accelerate at a blistering pace – with businesses, for example, able to offer increasingly sophisticated chatbots to support customer services.
The toolkit is a TensorFlow-based, modular framework designed to make it easier for users to create sophisticated conversational AI training configurations, using neural networks that involve new signal processing methods and deep learning algorithms.
As the company explains: “One of the major bottlenecks in training conversational systems is a lack of conversational data that captures the richness of information present in the abundance of non-conversational data that exists in the world.
“We therefore need good tools that can take advantage of the latter.
As Microsoft’s Vighnesh Leonardo Shiv explains: “To achieve… conversational scenarios where an AI may be required to adopt some persona with its own particular style and attributes, Icecaps allows researchers and developers to train multi-persona conversation systems on multi-speaker data using personality embeddings.”
The release comes as Gartner estimates that by 2022, 70 percent of white-collar workers will interact with conversational platforms on a daily basis, although a crowded vendor marketplace includes many unable to offer enterprise-grade solutions.
“There has been a more than 160 percent increase in client interest around implementing chatbots and associated technologies in 2018 from previous years,” said Van Baker, VP analyst at Gartner in a recent paper. “This increase has been driven by customer service, knowledge management and user support.”
Shiv notes: “At Icecaps’ core is a flexible multi-task learning paradigm. In multi-task learning, a subset of parameters is shared among multiple tasks so those tasks can make use of shared feature representations. For example, this technique has been used in conversational modeling to combine general conversational data with unpaired utterances; by pairing a conversational model with an autoencoder that shares its decoder, one can use the unpaired data to personalize the conversational model.
“Personality embeddings work similarly to word embeddings; just as we learn an embedding for each word to describe how words relate to each other within a latent word space, we can learn an embedding per speaker from a multi-speaker dataset to describe a latent personality space. Multi-persona encoder-decoder models provide the decoder a personality embedding alongside word embeddings to condition the decoded response on the selected personality.
In the coming months, Microsoft said that it will equip Icecaps with pre-trained conversational AI models that researchers and developers can either use directly out of the box or quickly adapt to new scenarios by bootstrapping their own systems.
Its codebase can be found on GitHub.