With over 360,000 registered Azure Bot Service developers, we’ve seen significant growth in bots and virtual assistants built on Azure. A major trend we’re following is the growing need for these assistants to support voice-first conversational experiences. As a result, we’re taking steps to make it even easier for developers to build virtual assistants with our virtual assistant solution accelerator and to add speech to their conversational applications with Azure Bot Service.
At this year’s Microsoft Build conference, we announced signup availability of the Direct Line Speech channel, which simplifies the creation of end-to-end solutions for voice-first conversational experiences. Today, we’re happy to share that the Direct Line Speech channel is now in preview for any developer with no additional signup or approval required. With this release, the Direct Line Speech channel has also significantly expanded its region support to enable faster and more reliable conversational experiences worldwide.
About Direct Line Speech
Direct Line Speech is a new channel that simplifies the creation of end-to-end solutions for voice-in and voice-out natural user interfaces with a few key components:
- An on-device API, available as part of the Speech SDK, simplifies speech and real-time supplementary signal communication to and from a compatible bot.
- A cloud service that coordinates wake word verification, speech-to-text, and text-to-speech capabilities for use with your conversational experience.
- A new Microsoft Bot Framework channel optimized for low-latency, high-reliability communication in voice-first scenarios.
The Direct Line Speech channel is designed to allow deep customization of your virtual assistant or conversational experience to fit your requirements and your brand, including the use of custom wake words to give your assistant its own unique name.
The Direct Line Speech channel is available for preview use today. Use an existing bot or create a new one. Visit the Azure portal to connect it to the Direct Line Speech channel, then get started with a device application for access to a variety of platforms and programming languages available. We’re excited to see what you can create and eager to hear your feedback.