Blog Post

Twilio Launches Roll-Your-Own Google Voice

Twilio, a San Francisco-based startup whose platform and services marry voice to the web world, is launching OpenVBX, an open-source web application that is in essence a roll-your-own Google Voice. With it, you get toll-free and local phone numbers, which can in turn be used to route calls to existing cellphones and landlines.

OpenVBX is not the first open-source telephony offering. Asterisk and Freeswitch are the most well known, but require a certain level of sophistication in order to be deployed inside corporations. Implementing OpenVBX is simple and yields the one thing users want most: a voice mail box that also forwards calls to different numbers.

Such simplification of VoIP-based services is a larger trend, other examples of which include Calliflower, a mobile-centric conference calling service. Earlier this month, Voxeo’s Tropo introduced OpenVoice, a virtual number application that can forward calls, handle voice mails (with transcriptions), send and receive SMS and make outbound calls.

OpenVBX is targeted at small- and medium-sized businesses. It’s multi-user, there’s an easy drag ‘n drop UI for building business applications, and a voicemail and SMS interface for managing communications. Says the company:

As an open-source application, software developers, IT departments and consultants can download the source code and freely customize it for any company, vertical or industry. Web developers can tailor OpenVBX by building plugins to automate and direct phone calls, such as a shipment status integration for e-commerce companies, a store locator for retailers, or appointment booking features for doctors offices.

In addition, OpenVBX provides Application Programming Interfaces (APIs) for deep integration with existing CRM and ERP workflows, and a theming system to rebrand the interface for clients. Software developers, IT departments and consultants can download the developer preview release of OpenVBX from to install on their servers.

Unofficial plugins have been developed for 37Signals’ Highrise CRM and Zendesk. An unofficial Foursquare plugin developed by Andrew Watson is the most interesting of them all: thanks to this plugin, when he is checking in at home, all his work calls go straight to voice mail.

So what’s in it for Twilio? Well, the system uses the underlying Twilio infrastructure, which means users buy phone numbers and connectivity from the company. I’m still playing around with a demo version Twilio shared with us — I’ll update the post when I’m done.

23 Responses to “Twilio Launches Roll-Your-Own Google Voice”

  1. Hello, I developed OpenVoice (OV) with the vision for it to be fully open-source by integrating with FreeSwitch (FS) and Asterisk. Here is the blog entry where I announced OV: An OV user will be able to select which backend service (Tropo, FS, etc.) to use. Of course some capabilities (such as IM) are platform specific but all backends will support call control required by OV. For developers, they will be able to use the same set of APIs to write client applications (mobile or desktop).

  2. I’m the founder of The name, OpenVBX, is a little misleading as only the PHP web application that let’s you administer the PBX is open-sourced while the platform is designed to only run exclusively on Twilio’s services at $.03-$.05 per minute – UNTIL NOW! By releasing OpenVBX, Twilio helped Teleku more than they currently realize. We have always supported the Twilio Markup Language (TwiML). We just needed a good test application that had extensive coverage of the Twilio API to finish our ability to test Teleku’s support for Twilio’s API and TwiML (in addition to our own easier API and PhoneML).

    Teleku’s implementation of the Twilio API and TwiML has been enhanced and now more than 90% of OpenVBX’s features running on Teleku with no changes to the OpenVBX code base except for changing the URL of the API from to This means that OpenVBX is now – well “open”. Customers are no longer “locked” into Twilio hosting to use OpenVBX. Since Teleku ultimately translates PhoneML and TwilML to VoiceXML, OpenVBX now runs on any carrier’s platform such as Voxeo!

    For the good of the telecom industry, we hope that Twilio will allow Teleku to contribute code to this exciting new open source initiative, OpenVBX! We believe that for an application to be considered truly “open”, the customer should not be “locked” into a vendor’s platform. Teleku solves this problem by allowing developers to run ALL of their Twilio applications on any carrier they desire as well as by using lower-cost SIP alternatives to deliver game changing telecom solutions.

  3. Ravi

    I’m the CEO of Voxeo. I wanted to comment on one of your points:

    “Can these startups survive based on the revenue generated by these apps?”

    Twilio, IfByPhone, and TringMe are startups and its fair to question if they can survive. Unfortunately far less than half of all startups survive. Most that do survive are consumed by larger companies that discontinue or significantly alter their products.

    I hope these startups do survive – they bring competition that drives innovation. It drives Voxeo to create even better products and services. Many of the ideas Voxeo innovated over the last ten years have been adopted by our startup competitors. Voxeo will adopt many of their new ideas back. Ultimately all customers and developers benefit.

    Spiritually Voxeo is a startup. We innovate and have agility like a startup. In reality we’re not a startup.

    Voxeo focuses on our customers and products, not on our competitors. Frankly we’re not very good at marketing our hyping ourselves. For example, most people don’t know that:

    • Voxeo is a ten-year old company, profitable since 2005

    • Voxeo has invested over $40m into our network and hosting sites

    • Voxeo runs 7 in-house data centers in the US, EU, and Asia

    • Voxeo does not use EC2 or any equivalent – we control our infrastructure

    • Voxeo owns all its platform technology (we don’t use Asterisk, etc)

    • All of our technology is “Unlocked”: 100% open standards or open source

    • Voxeo has just under 150 employees in the US, EU, and Asia

    • Voxeo has nearly 150,000 registered developers

    • Voxeo had 158 developers sign up yesterday (without a new product launch)

    • Voxeo will likely pass 200,000 registered developers this year

    In short Voxeo is stable, profitable, rapidly growing, and we love startups and developers.

    We started Voxeo ten years ago with a mission to make telephony apps easy for anyone to create and deploy. No one should be surprised to learn that after a ten years much of our revenue growth comes from larger software, enterprise, and carrier customers including Adobe, IKEA, Microsoft, Orange, SAP, and T-Mobile.

    However, at the same time voxeo is run by a “Coding Executive Officer”. I wrote our original cloud telephony platform in 1999. More recently I wrote most of the sample apps for our Tropo service. I’m not a typical MBA CEO. Voxeo has helped thousands of developers and startups grow their business plans into successfull realities. In turn, their success has helped us grow. These are the reasons why we invest so much of our profit into more free and new developer services than anyone in our industry. It’s in our DNA, it’s our passion — and its profitable.

    So what happens when startups do innovate far ahead of what we’re doing? Great! We also use our profits to drive acquisitions. We’ve acquired eight companies and product lines in the last four years. Entrepreneurs like to be part of Voxeo’s culture. Unlike others we work hard to keep the people, products, customers, and innovation acquired startups bring to us.


    -CEO, Voxeo

    Anyone can learn more about Voxeo products (and some of our acquisitions) at these sites:

    • @ Jonathan Taylor
      Thanks for the information. I have been following Voxeo and I know you guys are leading the pack when it comes to Telco Platform/Cloud Telephony. From a revenue and profitable standpoint, Voxeo is a classic example for other small startups that have great ideas but don’t know how to monetize the service.

      I guess my skepticism stems from the fact that for a startup to survive revenue stream plays a pivotal role. Iam saying based on my experience working in many telecom startups for more than a decade. Some of them closed shop even though we had a kick butt product. It all boils down to sustainable revenue model.

      Having said that, iam thrilled to see innovation happening in the Telephony arena. There is so much of potential in this space. Iam sure we will see lot more collaboration in future, which might even include operators joining hands with some of these startups to unravel new businesses.

      @David F. Skoll
      I agree with you on the reliability. Telco providers have spent billions or dollars on high availablaity,reliability and scalability. I can’t imagine what would happen if your landline goes down and you are in an emergency situation. It’s not the same as twitter/facebook or some web server going down. The repercussions are huge when it comes to service availability for real-time communication. Unless Service reliability is addressed by these cloud telephony providers, I don’t think it will be adopted by mainstream folks.

  4. @Ravi

    I like your comment on the group com offering from Verizon wireless. We have been working on a similar solution for a year, since we are tired of not being able to broadcast messages within closed user groups in a simple way. It’s still in alpha mode, but not for much longer..

    [email protected]

  5. Le Beau

    @David wrote:

    I’m not sure that “Cloud Telephony” makes sense. It’s all very cool and whiz-bang, but it comes without the reliability guarantees of traditional telecom service. Quick quiz: When’s the last time your POTS line was down? Now how about the last time your Internet access was down, or an EC2 instance rebooted, or something weird happened with IP routing?

    Very well stated. Here’s the problem — Om Malik is going to have a hard time addressing your comments objectively because he and his colleagues are going to be putting on stage the upcoming Structure conference which is all about the cloud. Because the cloud is sexy these days, a lot of people will take an all-or-none approach and think the cloud is the be-all end-all. Truth be told, these are early days for the cloud computing platform. Way early. As in a decade of development ahead of us (its not as if the cloud is here today gone tomorrow gold-rush mentality). Hence, all the more reason why your “roll your own” with Asterisk and POTS has great wisdom for today’s context. But in the high tech world and the bubble that Silicon Valleyites tend to live in, they often forget about wisdom! note: this isn’t to say we won’t see improvement in the decade ahead as the cloud matures. For most businesses it is high risk to depend on the youthful unproven startups in this space, but nonetheless its very admirable that startups like Twilio are pushing the envelope and trying (which is true to the entrepreneurial spirit)! Hence, you should also keep this in perspective and not be too hard-nosed about roll-your-own Asterisk on POTS!

    • @Le Beau, I agree with you. Cloud computing has enormous potential where it’s applied properly. It’s great for things like CRM, mail filtering, compute clusters, Web servers that can be dynamically provisioned, storage for backup, etc…

      I have a harder time seeing how it can apply to real-time high-reliability situations, at least until home and business Internet access comes with the same kind of service-level agreements as traditional telecom.

  6. “Can these startups survive based on the revenue generated by these apps?”

    I doubt it. I’m not sure that “Cloud Telephony” makes sense. It’s all very cool and whiz-bang, but it comes without the reliability guarantees of traditional telecom service. Quick quiz: When’s the last time your POTS line was down? Now how about the last time your Internet access was down, or an EC2 instance rebooted, or something weird happened with IP routing?

    Yeah, I thought so.

    We use Asterisk extensively and have all kinds of whiz-bang integration features; I have slides at

    It was easy and reliable because everything is in-house and our actual phone lines are reliable POTS lines. I’d never trust the Cloud for phone service; the Internet is way too flaky.

  7. This is something Asterisk or Freeswitch guys should have done long time back. Losers!. Companies like Twilio,Voxeo,ifByPhone and Tringme are making the Cloud telephony a reality. The success of some of these startups will bring in bigger player, and in the process some of these guys will be acquired or booted. There was time when creating anything in the Telecom domain was a humungous task. Not anymore. Some of the telephony servers like Asterisk can be hosted on Amazon servers. So what we are seeing is a Pre IMS type of scenario where in application servers are sitting in the cloud and are used for processing calls,sms,messaging etc. I’m sure operators will have a bigger role to play in the future of Cloud Telephony. Since twilio and others are not into building their own Telco network like magicjack, they still have dependency on third-party network provider for wholesale voice/SMS/messaging. Here is where operators can play a bigger role.

    All that said, the success of these startups depend on the developers and the kind of apps they develop using these platforms. These startups make money via the usage of Voice/SMS/Messaging from the apps. So the big questions is- are we there yet?
    Can these startups survive based on the revenue generated by these apps? In my opinion this will decide the future of Cloud Telephony API

    Om if you get a chance please take a look at Verizon Group Communications service that includes one number conference calling, Group messaging and Voice messaging

    Would love to hear your feedback on the service. Thanks

  8. @Adam Kalsey, thanks for the clarification. I have no problems with companies selling non-open-source products or services (my own company does that too.) But I do have a problem with companies that are misleading about what they do, and I think it’s misleading to call something an “open-source telephony offering” unless it really IS a standalone, open-source telephony solution.

    Note that Om Malik used the phrase “open-source telephony offering”. I have no idea whether or not Twilio represents itself that way.

  9. Steve hamilton

    David –

    I can’t agree with your point more.  
    Ultimately nothing is free or open about OpenVBX.  
    OpenVBX runs on a single proprietary, closed, pay-by-the-minute-for-all-eternity platform from Twilio.  
    OpenVBX supports none of the open voice standards (CCXML and SIP Servlets for example) or open source de facto standards (Asterisk AGI for example) that have been created over the last decade.

    Reality is you can do whatever you want with OpenVBX… as long as it makes Twilio money.

    In contrast Voxeo’s platforms are all either open standards, open source, or both.  Anything built on Voxeo platforms can be run elsewhere.  You are never locked in.

    For example:

    Like Twilio, Voxeo Tropo offers an incredibly easy to use but ultimately proprietary API for telephony.  
    Unlike Twilio, the Tropo platform itself is open source.  You can download the Tropo source and run it on any of ~5 platforms (including two open source platforms) that support the Java SIP Servlet standard.

    Companies like Twilio are embezzling the value of open source and open standards.   They create a facade of doing right for the cause — but ultimately they are only using that facade to create lock in and drive revenue to Twilio.  Twilio and their kin do not help advance either open-source or open-standards.

    • David F. Skoll

      @Steve Hamilton: I didn’t intend my comment to spur cloud vendors to take shots at each other. I’m not sure Voxeo is any more an open-source telephony platform than Twilio.

      I’d love to be corrected, but as far as I know, Asterisk, FreeSWITCH, Yate and GNU SIP Witch are the only truly open-source telephony platforms available, in the sense that I can download some open-source software and actually use it to do telephony without any external dependencies.

      • David,

        Steve isn’t a Voxeo employee. I am, however.

        You’re correct, Voxeo’s products are not open source. We do, however go to great lengths to ensure you’re never locked into our platforms. We either fully comply to existing open standards (like VoiceXML) or where no standard exists, we open source the bits you’d need in order to take your app elsewhere.

        We want you to stay with us as a customer because we’re awesome, not because you have no other choice.

      • Whoops, a typo in my previous comment changed the meaning pretty dramatically.

        It should have read “Voxeo’s products are not ALL open source.”

        Voxeo has a great number of open source offerings, from our SIP and XMPP framework Moho to the Tropo platform itself.

    • Interesting points — I hadn’t thought of that. I (obviously) am a big fan of hosted service running GPL code model, as that is how WordPress and work. I’ll take another look at Voxeo.

  10. Om, thanks for the mention of the OpenVoice app built on top of Tropo ( ). OpenVoice is open source, written in Ruby for Rails and available for download from Github. It’s a great example of the kind of apps our developer community are building on top of Tropo. With Tropo, developers are building apps that use voice, complete with automatic speech recognition in 8 languages, and also use SMS, instant messaging and even Twitter for interacting with users. Developers also get phone numbers available around the world, Skype and SIP connections and more.

    We were pleased by the response to OpenVoice at Google I/O last month and we’re constantly amazed by the new apps people are building on top of Tropo.

  11. David F. Skoll

    You wrote: “OpenVBX is not the first open-source telephony offering.” That’s true, because OpenVBX is not a telephony offering at all. You can’t compare it to Asterisk or Freeswitch which actually do telephony. OpenVBX is really just a client library for Twilio’s APIs, which are not free and not open-source.

  12. Charlie B

    “As an open-source application, software developers, IT departments and consultants can download..”

    I hope their software is done with more care than their copy writing. Look up “dangling participle”, guys.