Build AI-assisted calling workflows with Azure AI Speech and Azure Communication Services
Published date: 06 September, 2023
As a modular communications platform for businesses, our goal is to help developers not only implement programmable workflows across their applications, websites, and mobile platforms, but also enhance and improve customer outcomes within those workflows.
Now available in public preview, developers have the ability to add speech capabilities into Azure Communication Services call automation workflows with Azure AI Speech. Azure AI Speech enables powerful conversational AI experiences for customers by recognizing specific phrases as well as free-form sentences spoken by a user like – “hey, my car broke down, I need help;” enabling businesses to adapt to changing customer needs in real-time and shorten the time spent navigating Interactive Voice Response (IVR) menus.
With features like Speech-to-Text and Text-to-Speech, businesses can also enable virtual agents to have voice interactions in over 100 languages and use neural voices to make interactions more natural and engaging. Or enhance virtual agents by adding emotions using the Speech Synthesis Markup Language (SSML) feature. These speech capabilities also simplify the use of Azure Open AI Service in calling workflows with just a few lines of code. Businesses can feed ongoing interactions to the Azure OpenAI Service to identify intent and personalize recommendations based on user activity and sentiment. Businesses can also analyze historical interactions to identify trends that help optimize their operations, garnering actionable insights to improve future customer outcomes.
Walk through the sample code here