Here at Dolby.io, we believe in creating products and services that reduce complexity and create better end-user experiences. Towards that end, over the past 12 months, our product and engineering teams have been exploring ways to increase the scale and quality of applications built using our Communications WebSDK. Specifically, asking how we can enable developers to leverage more technologies that Dolby has developed in its near 60 years of research in sound and signal processing.
Earlier this month, Dolby.io completed this year-long journey and successfully launched its Communications WebSDK 3.5, featuring an entirely new technology stack built on WebAudio, WebAssembly, and creative use of foundational WebRTC components.
Through this new SDK, Dolby.io can offer its Dolby Voice Codec that improves conference scalability and introduces new media processing capabilities for developers who build browser-based applications.
The Dolby Voice Codec was previously only supported in Native Desktop and Mobile applications.
About Dolby Voice
Dolby Voice is an award-winning technology that enables exceptional audio experiences in real-time communications applications. Dolby Voice captures how things sound in natural environments and leverages signal processing to enhance conversations across the web and within desktop and mobile applications. These enhancements include suppressing unwanted background noises, leveling mic volumes in multi-party conference calls, and improving dialogue intelligibility through Spatial Audio. Dolby Voice enables more productive business meetings with your colleagues, enjoyable virtual happy hours with your friends, and better health outcomes through online consultations with doctors.
Dolby Voice features and benefits
|Dynamic Audio Leveling||Dolby Voice automatically levels microphones in multi-party calls, balancing speaking and listening volumes across participants. The result is a natural-sounding conversation.|
|Noise Reduction||Dolby Voice automatically removes unwanted background noise from your call by suppressing distracting sounds like barking dogs, street noise, or keyboard clicks.|
|Echo Cancellation||Dolby Voice cleans up your audio stream by removing the conference output from your mic, ensuring others don’t hear themselves echoing back.|
|Spatial Audio with Manual Placement||Hear what you see. Spatial Audio enables position-based listening experiences that make it easy to identify who is talking in a multi-party video call or 2D and 3D environments. If someone on the left side of your screen talks, their voice will come from the left.|
|Dolby-managed end-to-end audio experience||Instead of handing off audio to a 3rd party service, Dolby.io manages the end-to-end audio experience for our customers. Dolby.io captures the audio from your microphone, enhances it using Dolby Voice technology, and delivers it across the web to its destination. Dolby.io leverages custom codecs and algorithms to fix corruption due to bad networks, and our backend is designed to optimize the overall conference experience. A Dolby Voice conference is characteristically “calmer”, as we actively manage the amount of content that gets into a call.|
Dolby Voice in action
The below images and audio clips demonstrate how Dolby.io improves conversational speech in loud environments.
Raw audio capture spectrogram
The top image is a spectrogram showing audio captured at a microphone – conversational speech in a restaurant setting. You can see the high-energy voice components in yellow popping out from the sea of red background noise.
WebRTC processing spectrogram
Traditional processing in WebRTC works to remove some of that noise, as shown in the second image. However, once a couple of people join a call with that much noise, everyone will need to start riding the mute button.
Dolby Voice spectrogram
In contrast, Dolby Voice uses artificial intelligence to identify the frequency components belonging to the foreground speech and works to remove everything in the background. The result is a more immersive experience, enabling users to concentrate on the conversation rather than managing the technology.
Dolby.io does not charge extra to leverage Dolby Voice. It is a core technology within the Dolby.io Communications APIs and something that differentiates Dolby.io from other WebRTC API platforms.
Check out our pricing page for more information.
Developers, product owners, and business teams can benefit from decades of Dolby research and development to build high-fidelity, immersive real-time audio and video experiences that exceed their customers’ expectations by leveraging Dolby Voice within the Dolby.io Communications APIs.
Contact sales or sign up for a Dolby.io account and start building with our updated Web Client SDK!