Amazon Lex now supports 8 kHz telephony audio for increased speech recognition accuracy
The Amazon Lex speech recognition engine has now been trained on telephony audio (8 kHz) to provide increased speech recognition accuracy for conversations over the phone. Amazon Lex enables your users to interact with your application via natural conversation using the same deep learning technology as Amazon Alexa to fulfill most common requests. Amazon Lex chatbots maintain context and manage the dialogue, dynamically adjusting the responses based on the conversation.
For example, Amazon Connect, a simple-to-use cloud-based contact center service that scales to meet your needs, uses Amazon Lex chatbots to handle first level customer support, and can solve many customer problems without involving a human operator. When necessary, Amazon Lex can transfer a customer support call to an agent, with full context. Developers can also integrate Amazon Lex with the contact center technology of their choice using the Amazon Lex APIs. Amazon Lex makes setting up a natural language experience simple, fast and self-service, at a fraction of the cost of other solutions.
Amazon Lex with 8 kHz telephony audio is available today in the US East (N. Virginia) AWS Region. For more information on using Amazon Lex with your Amazon Connect contact center, visit the Amazon Connect product page. Visit the Amazon Lex product page anytime for details on Amazon Lex.