Announcing Video

Announcing Video

Howdy, there’s a whole lot that’s new in the Video world. Check out our latest updates and ships here.

Today, we announced Video. Developers can now create fundamentally better experiences by embedding live peer-to-peer video into mobile and web applications. Video is much more than just being able to talk face to face. Video enables users to share what it is they’re working on, and to express things that cannot be expressed with voice and text alone.

Attendees at Signal will be the first to get their hands on the beta (May 19 & 20 in San Francisco). In the meantime, you can get on the waitlist for early access on www.twilio.com/video and learn more at our meetups in New York (Apr 20), San Francisco (Apr 21) and Mountain View (Apr 22).

Using video to create better experiences

Before we jump into the details, here’s an example of why we’re excited about this. Zendesk was one of our first customers to take Twilio Video for a test drive. The Twilio crew flew out to Zendesk’s office in Dublin, Ireland to build the proof-of-concept. The idea was simple: What if we could take advantage of the camera on a mobile phone to completely change the mobile customer service experience? Zendesk already enables customer support teams to answer customer service phone calls directly in the Zendesk browser app. They also enable businesses to embed Zendesk’s customer support features directly into customer-facing mobile apps through their Embeddables SDK. We wanted to combined Embeddables, Zendesk’s agent interface, and Twilio Video to help customer service reps see and solve problems first-hand.

The combined team had a prototype integrated into Zendesk working within a day, and a polished demo within a week. In the final demo app, customers can tap a button to connect with an agent over video, and share the front or rear cameras from their mobile, or the screen of their mobile device. Agents can capture video frames and add them directly into a Zendesk support ticket when they see something of interest. The outcome is a richer customer support conversation, one where a customer can show a support rep the specifics of the problem they are grappling with, rather than trying to explain it in words.

Twilio Video is a combination of cloud infrastructure and client-side SDKs. The Cloud Infrastructure provides all of the orchestration for setting up a video call: Registration, signaling, and media routing. The new Twilio web and mobile SDKs provide the other half, making it really easy to consume these services across the different supported platforms of iOS, Android and JavaScript.

Built on WebRTC

When we set out to build Video, we knew we wanted to one day reach every device and every platform with realtime communications capabilities. This is why WebRTC, the open standard for realtime communications, sits at the core of Twilio Video. WebRTC already has strong support in the Chrome and Firefox web browsers, allowing users to leverage video capabilities without installing software or plugins. The open source WebRTC libraries are designed for cross-platform compatibility, opening up a world of mobile, desktop and IoT devices to the standard.

But WebRTC is just a piece of the puzzle, not a solution. To take advantage of the standard, you still need to provide registration, signaling, and network traversal to enable endpoints to find one another and establish connections. Twilio Video makes this happen with its Video orchestration services available in 28 data centers in 7 geographies around the world.

The cloud isn’t the only place where some heavy lifting is required. WebRTC, in its native form, is optimized to work well in a web browser running on a powerful desktop device. Twilio’s SDK team is investing massive energy to make sure the platform performs well in more constrained computing platforms, such as iOS and Android devices, and on a variety of hardware configurations. Optimizing media quality and SDK performance is an ongoing mission–we’re always testing and tweaking to squeeze the best performance out of the platform.

Designed for a new breed of applications

We know that the next generation of communications applications will change the way we connect, work together and share with one another, and get things done. We know this new world needs a new breed of communications primitives.

That’s why Twilio Video has been designed from the start to support multi-party voice and video calls. Up to 4-way peer-to-peer calling with Twilio Video gives you the ability to add participants ad-hoc to a voice and/or video conversation. To enable that, Twilio Video provides:

• Unlimited video connections: Registration and Signaling capabilities are globally distributed and designed to scale elastically as load increases.
• Low latency relay: Automated network traversal enables peer-to-peer connections across firewalls. The closest Twilio data center is automatically chosen as the media relay point when firewall conditions prevent a peer-to-peer connection.
• Optimized Mobile Media Stack: Optimized for iOS and Android, Twilio SDKs ensure great quality real-time media on mobile devices despite common network challenges such as jitter, packet loss and latency
• Cross Platform Data Channel: All peer to peer calls have access to a DataChannel between users, allowing data transfer directly over DTLS/SCTP alongside the media flows. This can be used to send files, send arbitrary JSON data or anything else. For example, a user can draw on top of the video they’re viewing, and the data for that can be transferred and shown immediately on the other side.

We’re really excited about Twilio Video. Whether you’re building a face to face communication service, or building video into your existing app to create great customer experiences, we can’t wait to see what you build.