The Developer Challenge, and Flash audio players

May 1, 2008

Apologies for the long time since posting… I guess travel and jury duty will do that.

Two things: The Developer Challenge entries have closed, and I think we have a very exciting batch of participants. They really show off the wide range of ways in which MyVox can be applied to solve problems and create new products. Now we have the hard problem of picking winners. Once that’s done, you’ll learn more about all the participating applications.

Secondly, it’s very common for MyVox apps to need a way to play back their audio after it’s created. While that’s technically not our business, we like to help our developers out with any reoccurring situations, and this definitely qualifies. To that end, we’ve put together a set of MyVox-branded audio player widgets, and have also put our stamp of approval on some freeware widgets that we think work well. Ours are here:

http://api.myvox.com/tools/players.jsp

And the freeware ones are here:

http://musicplayer.sourceforge.net/

They are pretty easy to use, but if you’d like more information, check out the MyVox Reference Library, where permanent docs explaining the use of the Flash players will show up momentarily.


MashupCamp VoiceMap

March 20, 2008

At MashupCamp speedgeeking, I asked the people making the rounds to participate by putting a map point on their hometowns, and then calling in and recording themselves saying their names and what they thought of MashupCamp.  Here’s the result.  (Most people liked MashupCamp.)


MashupCamp so far

March 20, 2008

MashupCamp has been a great experience, and we still have a day to go. It’s a special experience getting to share a room with a bunch of like-minded geeks looking to investigate a topic or take on a challenge. I’ve been fortunate enough to attend three unconferences this past month; MashupCamp has been the best of them.

I both led a session and “speedgeeked” yesterday. Speedgeeking is like speed dating, except you are demoing your app instead of yourself. I showed off the VoiceMaps app from our gallery, having each visitor add a point on the map with his or her hometown, and then recording that person saying something about who they are and what they think of MashupCamp. You can see the result here. Lots of fun, and also exhausting… five minutes sounds like a long time to talk to somebody about your app, but in reality it was never, ever enough.

The session was good too. I focused less specifically on MyVox and more on sharing what I knew about the tools and approaches available to folks who want to incorporate telephony or voice into their apps. Most of the folks here have never thought about telephony and how it fits into their development world, and it seemed like a number of folks got a lot out of the experience. While I was hoping that there would be folks from Ribbit around to comment on their take on APIs and voice mashups, some attendees of the MashupUniversity were able to relay what they learned from Ribbit’s presentations there. You can see my hasty notes on the session on the MashupCamp wiki.

Coolest app seen so far:  Yahoo’s FireEagle.  Still have to delve deeper into this one, but as somebody who’s been trying to figure out how best to bring location-based services into the mix of things we work with, this looks exciting.


Gallery, press, and new feature roadmap

March 14, 2008

For those who have not heard, the MyVox Gallery site is now in place.  This is where we feature all manner of apps that combine phone and screen (be that your laptop screen or your mobile screen) in cool ways.  You can check it out at http://gallery.myvox.com.

We’ve gotten a lot of press over the past week as we’ve done the really-truly-official launch of MyVox out to the world.  Here’s at least some of it:

Some MyVox developers have asked me for a roadmap of upcoming feature releases.  Here are some of the things we plan to release soon, though I can’t give you an exact schedule:

  • Starting with the phone:  Right now, all MyVox interactions start with the user in front of some sort of screen, so that you can get your phone number and PIN.  However, there are a lot of great apps that become possible if the user can begin interacting while away from the screen - so we will enable that soon.  More details to come.
  • Dedicated phone numbers:  If you need your own phone number for people to call into, we’ll be able to set you up with one.  This is also important for making the above feature work well.
  • Emailing audio files:  You’ll be able to have MyVox recordings go to an email address instead of being stored on our server or uploaded to a URL.  This is mostly a convenience feature; there are plenty of ways for developers to do this themselves, but we thought it would be nice to build it into the API.
  • Podcasting:  When you think about it, a RecordingList is very similar to a podcast feed.  In both cases, you’ve got a set of recordings that have been made at different dates and times, and which other people may want to consume.  Soon you’ll be able to get RecordingLists with the appropriate RSS formatting to allow podcasting directly from MyVox.  This will be great with apps like VoiceBlog.
  • Caller ID:  Sometimes it’s nice for an app to be able to access the phone number of the caller, so we’ll be making that available soon as well.

Got another feature you want to see?  Let me know in the comments.


Upcoming conferences

February 28, 2008

Looking to hear more about MyVox?  If you are attending one of these conferences, or will just be in the area when I am, look me up and we’ll chat:


The growing world of voice APIs

February 25, 2008

I remember when the phrase “voice mashup” was one we only heard in our offices, talking to each other. Now, it feels like I can’t go a day without reading “voice mashup” or “VoIP mashup” or “phone mashup” somewhere.

And in conjunction with that talk, there’s also a lot of activity - many companies recognizing that voice and phones can become serious tools for serious application development, if they can be offered up in a form that lets developers from other realms make use of them. So we are seeing this… well, not explosion, but at least a burp of activity around voice- and phone-related APIs. Here are a few of the players:

* Cepstral:  Long-time providers of text-to-speech, Cepstral recently launched VoiceForge, which offers text-to-speech through a SOAP-based API. You can pick from about thirty voices, and even train VoiceForge to produce your own voice from your text! Cepstral prices their

* Voxbone:  Providers of inbound VoIP lines and phone numbers, Voxbone has an API to allow anybody to sell and provision Voxbone lines.  Use of the API itself is free; Voxbone makes its money on the sales of lines.

* Evoca:  Evoca has an API with similar capabilities to that of MyVox.  However, use of their API is not free; it is designed for use by business looking to streamline the development of certain kinds of voice mashups.

* Ifbyphone:  Ifbyphone offers an API for building “voice forms”.  This is basically a way of configuring an IVR to collect information from a user and deliver it to another application.  Ifbyphone’s API is not free - like Evoca, the target is businesses who are willing to lay out cash to get easy access to such capabilities.

Whether or not I agree with each company’s business model (and, to be clear, I like some of the above a lot), these companies, together with MyVox, demonstrate growing recognition of an unmet need.  That need is one that comes from developers of all sorts who want to harness the power of the phone, and tie it into their applications, but do not want to deploy, maintain, or learn the esoterica of phone technologies.


Breakfast with Jeff Pulver and friends

February 21, 2008

Jeff Pulver is hosting a series of open breakfasts with folks who want to sit and chat with him and other attendees. It’s a great concept: no speakers, no agenda, just the chance for conversation to take place naturally - something that breakfast is good for (except everybody is still waking up). I went to the Miami breakfast this morning.

We had about twelve people at our breakfast. Many people knew each other already, but everybody was happy to talk to those of us who were newbies as well. Some pics here: http://www.flickr.com/photos/adc/tags/smbmiami/show/ Good, creative, entrepreneurial folks - a real find in the Miami area (which is a bit dry of technophiles compared to, say, Boston, New York, or the Bay Area).

It’s an interesting challenge trying to get the word out about the MyVox API. How do you best reach developers, and show them you have something truly useful? Developers are typically very busy, skeptical folks who typically get a lot of their information through outlets that set a very high bar… precisely because they cater to skeptical, busy people. It can be difficult to demonstrate relevance when you are relatively unknown (as opposed to an API from, say, Flickr, or Facebook, or Google). There are many answers to how we are trying to tackle this problem, but breakfasts like this represent one angle; a better version of the same is BarCamps, where you have a concentrated group of developers who are there precisely to learn and exchange ideas.

Speaking of, I will be attending BarCampMiami on Feb. 28th, where MyVox is a sponsor. Will you be there? If so, seek me out!


Social Networks, Phones, and Context

January 28, 2008

Speaking on a panel at IT Expo on Tuesday got me wanting to put down in written form some thoughts on how social networks and VoIP go together.

For decades, the phone was the primary way in which people communicated with their “strong links”. Now, there’s another and very different one, and that is the social networking sites. The kinds of activities people engage in on these sites is wild and wildly varying, especially as sites like Facebook have offered development environments to allow third parties to go wild with integrated applications. But the closest thing to a common denominator is that everybody is engaging in most of these activities together. That’s why they are on the social network in the first place - they are playing games with each other, forming clubs, exchanging messages, and so on.

So the natural question, and one which others have asked before, is how social networks and phones should begin to go together? After all, if you use both to communicate with your friends, and if phones offer personality, convenience of communication, and immediacy where social networks offer variety, convenience of activity access, and , there intuitively ought to be ways to combine the respective merits of each in useful ways.

Some people are already exploring this. Alec Saunders and iotum have their Free Conference Call application for Facebook; at myVox, we have our Voxcall widget, which lets people see when a page owner is online and invite the owner to a call. These are good, important first steps, but I think we all have to agree that they are precisely that: tentative, exploratory probes of the surface of integration.

To get really deep, you need a stronger link, with rich information flow both directions, between the phone and the social network. And a key element of that is the availability of context. Here’s an example:

Say Allan is on Facebook, and he’s flirting with Bel. Programmatically, we can recognize this because Allan is interacting with Bel through an app that describes itself as a “dating” app, or because of the particular actions he’s taking within an app, or even because of the words he’s using. He and Bel decide that they want to talk to each other. Allan picks up the phone. Because the phone is aware of the context, it:

1) preemptively asks Allan if he’d like to call Bel in particular;

2) when he says yes, automatically determines that this is a dating-related call from the above context;

3) automatically masks Allan’s caller ID;

4) shares with Bel’s phone that Allan is calling, and the context;

5) shares with Bel’s phone any other personal information that Allan is comfortable passing on, like perhaps a userpic;

6) shares the same with the Facebook, so its applications can react to the fact that Allan is calling, should there be cause to do so, and perhaps also indicating that Allan is on the phone and not available for other calls.

What we have here is a phone interaction that begins to stitch the social network and phone together. It takes advantage of context, tailoring the phone experience accordingly, and then passing back information to the social network “operating system” from which the network’s apps can benefit. And it’s a pretty basic example, really.

But the current state of affairs with consumer VoIP (the most likely phone-based starting point) is one where we are very far from being able to entertain this sort of thing. On the social networking side, many of the elements are there, though not the sort of generalized and standardized context info that would really help phone services to “read the mind” of the user. The real challenge is with the phone service, and with developing a dynamic caller experience that begins the moment you pick up the handset.

nick AT voodoovox DOT com

GM, VoodooLabs


VoiceMaps (or “even I can do it”)

January 25, 2008

One of the key goals we set for ourselves in developing the myVox API was to make it easy enough for me to use. I used to do software engineering, but have no formal training, and tend to hack code together rather than really engineer it. And I haven’t programmed with any regularity for years. So that gives us the Nick Test: if it’s too hard for Nick to work with easily, then it’s not good enough.

I think this proves the team passed the Nick Test with flying colors. It’s a Google Maps/myVox mashup I put together that lets you create a “voice map”. Basically, you can add pushpins to the map, then record voice for each pushpin. Click on the map to create a pushpin, then click the pushpin to add voice. If you return to a pushpin that already has voice, the recording will still be there.

It’s definitely a demo, not a finished product. There are so many great things that could be added to something like this:

  • Persistence: If I leave the map and come back to it, the pushpins and recordings are gone
  • Instances: Different users should be able to create their own maps and save them individually
  • Photos: Combining photos and voice on a map would be very powerful
  • Text: I should be able to put up text comments too
  • Better controls: Right now, you can’t delete a pushpin from the page, or delete/overwrite a recording on an existing pushpin, and if you double-click to zoom, you end up also accidentally creating a pushpin
  • Prettification: The info windows are a little clunky right now, and a better Javascript programmer could produce a slicker recording experience

But the point is that I, as a rusty and inexpert coder, was able to throw this together in just two hours, without any experience with the Google Maps API. That’s a testament both to the myVox API and to Google’s.

Take a look and see what you think. Want to make it better? We’d love to see your work. If you make a big improvement, post it to the $25K Developer Challenge… you could earn yourself some cash, too.

nick AT voodoovox DOT com

GM, VoodooLabs


3..2..1..Launch!

January 25, 2008

Here at HQ, we’re all very excited about the launch of myVox - but more than that, we’re really curious. What will people do with this thing?

See, we made a bet with ourselves about six months ago. The bet was that the wide world of developers would build much cooler things with our voice-recording tech than we would ever create for ourselves. Part of this is the straightforward parallel-processing proposition - there are way more of you than us - but there’s also the fact that you probably think and care deeply about topics and applications that are completely beyond our experience. If you check out our embryonic gallery, you’ll see applications like Blabberize that we just would never, ever have thought to build ourselves… and yet, having seen it in action, we can only shake our heads and wish we’d been that creative.

There are more Blabberizes out there, and we think the thing that’s been holding back their development has been access to fast, free tools for doing the voice recording bit. Phones are hard, phones take work, phones take hardware, and even though there are ways to get it done (we do love Asterisk, to give an important example), many developers have better things to spend their time and money on.

At least, that’s our bet. I hope you will prove us right.

nick AT voodoovox DOT com

GM, VoodooLabs