Unless you've been hiding under a rock you know that WebRTC is posed to be the next big thing in real time communications.
If you're familiar with the technical details of WebRTC you also know that WebRTC doesn't mandate a signaling protocol - that's left up to well, whoever. For many of us coming from a SIP/telephony background it's made the most sense to use a signaling protocol we all know - SIP.
WebRTC makes extensive use of WebSockets and this combined with various JavaScript SIP libraries (sipml5, sipjs, jssip, etc) allows you to do call control using SIP from a browser to a remote system over WebSockets. This assumes, of course, that the remote system supports SIP over WebSocket transport. Then there's also the matter of implementing the various requirements for WebRTC media support such as ICE, STUN, DTLS, SRTP, etc. In short, implementing full WebRTC support is no small task.
Over a year ago I worked with the FreeSWITCH developers to get support for this approach to WebRTC added to FreeSWITCH. As usual, Tony and team delivered a very impressive result - the ability to use the aforementioned JavaScript libraries to call into a FreeSWITCH system using audio, video, or both - all from your browser! Of course once you were connected to a FreeSWITCH system all of the existing functionality was available to you. Conferencing, bridging to existing/legacy endpoints, etc. It wasn't a stretch at all to connect from Chrome using ICE, SRTP, OPUS, and SIP over secure WebSocket and then bridge to an endpoint using SIP over UDP and G729 (or even a PRI)! It boggles the mind to consider what is happening to the audio alone in this scenario - encrypting/decrypting, transcoding, and resampling. All magically and masterfully handled by FreeSWITCH.
One of the most interesting aspects of WebRTC is the ability to develop applications using the full power of the web and the browser environment. For years now we've seen WebRTC sample applications emerge demonstrating just how easy and powerful the peer-to-peer capabilities of WebRTC are. Some of these even demonstrate increasingly complicated "multi-peer" scenarios involving multiple parties.
Where does FreeSWITCH fit with all of these new-fangled web technologies? First, I've already mentioned the ability for FreeSWITCH to bridge to endpoints on different networks. Even though WebRTC shares some standards with these legacy endpoints, direct communication or even media will almost certainly never be possible. Requiring DTLS and SRTP virtually guarantees that.
FreeSWITCH also hosts many powerful applications - including voicemail, conferencing, and even entire custom applications written using LUA or Javascript. Custom applications can also be developed using ESL - a powerful event socket with accompanying library.
Historically there hasn't been a good way to make use of all of this functionality - the power of HTML5/JS/CSS in the browser (via WebRTC) and the power of voice/video applications hosted in the cloud using FreeSWITCH.
Until now.
Verto is an exciting new Javascript library and FreeSWITCH endpoint module. Together they allow web developers to use a single Javascript library for call control and FreeSWITCH event handling and interaction. That's right - call control and ESL in the same endpoint/protocol/library!
The FreeSWITCH endpoint module is configured to listen on a WebSocket or Secure WebSocket (or both). The Javascript library is included and configured to point to the FreeSWITCH instance. With a single library and simple API a web developer can make full use of a remote FreeSWITCH system using WebRTC within minutes!
One more thing - WebRTC (by way of OPUS) supports two channel audio (stereo). As part of Verto development the FreeSWITCH team decided it was finally time to implement stereo as well!
Make sure to stay tuned for more about WebRTC and Verto but in the meantime - enjoy the demo!
