Google’s chatbot ‘Meena’ aims to bring the human touch that others couldn’t

The team trained the chatbot with around 40 billion words and 341GB of text data which also includes social media conversations.

Google introduces chatbot ‘Meena’. (Pixabay)

With AI platforms like Alexa and Siri improving every passing day, the next natural step for tech firms is to make them more human-friendly, at least in terms of conversation. We saw Samsung attempting this recently with its 'Neon' platform and now Google is trying to do the same with its new open-domain chatbot named 'Meena'. This was announced by Daniel Adiwardana, Senior Research Engineer, and Thang Luong, Senior Research Scientist, Google Research, Brain Team in a blog post.

Meena Sensibleness and Specificity Average (SSA) compared with that of humans. (Google)

Meena Sensibleness and Specificity Average (SSA) compared with that of humans. (Google)

"Meena is an end-to-end, neural conversational model that learns to respond sensibly to a given conversational context. The training objective is to minimize perplexity, the uncertainty of predicting the next token (in this case, the next word in a conversation). At its heart lies the Evolved Transformer seq2seq architecture, a Transformer architecture discovered by evolutionary neural architecture search to improve perplexity," explains the blog post.

'Meena' is not really available for open-source as of now. But one can assume this to happen in the future. Google says that the chatbot has over 2.6 billion parameters, a key reason why it can respond to queries more naturally than other AI chatbots. Also mentioned is that the team trained the chatbot with around 40 billion words and 341GB of text data which also includes social media conversations. It also uses Google's Seq2seq model, which in simple terms, is a neural network that reads words placed next to each other in a paragraph and checks if the relation between those two makes sense.

Google's Meena chatbot in conversation. (Google)

Google's Meena chatbot in conversation. (Google)

Google researchers add that 'Meena' has a single evolved transformer encoder and 13 evolved transformer decoder blocks. What this means is that a single encoder to help it understand the conversation and 13 decoders that help it to form a reply.

The search giant has been lately making strides in the AI and human speech segment. A couple of years ago we saw Google demoing its Duplex platform and Google Assistant's ability to take calls on your behalf. And chatbot 'Meena' adds to it if nothing else. We can only wait and see when and where Google plans to use it.