Amazon Lex

Amazon Lex is a service for building conversational interfaces using voice and text, in exactly the same way the Alexa digital assistant provides an interface to compatible household devices.

Speak to an expert

Amazon Lex

Like Alexa, apps built with Lex can respond to user queries both with naturalistic replies and by initiating tasks and interfacing with other software and hardware – all of which can be specified and customised by you.

Responsive virtual agents

Chatbots powered by Lex can engage in complex and naturalistic conversations, answering queries and performing automatic tasks, then seamlessly connecting customers to the right human agent where escalation is required. All of this is enhanced by custom business-specific context, ensuring correct interpretations of user queries.

Interact through speech or text

Incorporating automatic speech recognition (ASR) and natural language understanding (NLU), Lex is capable of receiving queries via text or speech input, understanding the user intent behind them, and replying in appropriate manner. These capabilities are powered by the same technology as Alexa, streamlined and trained through millions of interactions with Amazon customers.

jakob-owens-dGXGDairmK0-unsplash - Softwire

Seamless and cost-effective scaling

The platform carries no upfront costs or minimum fees, charging only for the text or speech requests that are made. Usage scales up or down automatically based on need, so you’ll never be delayed by lack of capacity or pay for more than you use.

Our work with Lex

Softwire partnered with SentiraXR, developers of cutting-edge virtual reality simulations for training in the medical field. We developed a VR environment package which uses Lex to direct the simulation based on interpreted user utterances.

This application involved building a robust utterance-to-intent model based on real queries. We incorporated this model into other aspects of the package, allowing designers to base their work around identified user intents.