I was really impressed by the pitch from Tellme earlier today. (Although, despite them being great voice usability experts, their web site is a pile of Flash-infested crud.) I’d assumed IVR stuff was dull. On the basis you’ll get the best insights from the session that superficially is least attractive, I selected this session. Turns out it was a good choice.

I really liked the idea as telephony as the most intimate medium — a whisper in your ear. But what Tellme really have cracked is making the IVR experience much closer to interacting with a human, and not a string of audio files tacked together with some shell scripts. They played lots of examples of really, truly awful IVR experiences. And then what they did to them.

This is important because good experiences drive real business. The customer’s impression of your business and brand is derived right from the experience they have. The example Tellme gave was UPS. Their old IVR system was very, very slow. The messages went on and on, slowly read. Do you really want UPS to be associated with “slow”? Thought not. Most of all, a good experience creates trust in your brand.

The first thing they’ve cracked is making the voice experience more seamless. They’ve created vast libraries of all sorts of clever combinations of phrases which get blended together by cognitive psycholgy and linguistics experts. And the result is super-impressive.

Their voice libraries go beyond what’s known as “single prosody”, the old-style IVR where you heard broken-up phrases glued together like “departing | Saturday. | July. | 22nd.” Instead they have multiple prosody — “departing | Saturday, | July 22nd” etc. (note the comma after Saturday.) It works. But they’ve had to record over 37000 WAV files just to read back numbers!

They’ve also cracked “points of co-articulation.” You can’t record every possible combination of terms. So record the first term followed by an example second one starting with one of the 40 phonemes in English — “Hi John”. Then record all the possble second terms: “James”, “Jim”, etc. Then splice in the right second term just in place of the example one. Again, the result is impressive.

You really can tell the difference in terms of comprehension and memory retention.

They also did a great pitch on optimising the usability of IVR systems. The phone is a linear presentation, and taxes short-term memory. You don’t have a 2D screen with bold, drop-down boxes, etc. The boundaries are also invisible. It’s not like the Web. There’s a strong “recency effect” — the last thing said (“press 0 for operator”) is first thing remembered.

So they have a bag of tricks. Personalise. For example the sports team “squeaked by” if you support that side vs. “lost a close one to” if you don’t. They “instruct as you go”, deferring navigation instructions to the time they’re needed. (Lazy evaluation always deserved a comeback…) They use “progress markers” - “First, tell us…” “Next,” “Lastly”. Adopt colloquial language, not written English. Optimise to meet user goals, not sub-tasks. And so on.

I’ve glossed over a lot of ineresting detail, and good stuff. If only they could put up a few corporate blogs and share their cool innovations and work on an everyday basis!

Martin's other tales are told on his Telepocalypse blog.

An interesting panel on design and usability. Most interesting is BJ Fogg sharing his useablity thoughts behind YackPack. This is really a neat application. I'd heard about it but not followed it up in the past. In his case he's actually subscri bed to twenty six packs right now. The interface is very simple. He keep reinforcing taking things out that people aren't using. Basically Yack Pack is a fun method for broadcast audio messaging. Both voice and text are available. Check out the explanatory video. It's intriguing. Here's an example of how to send an audio blast to a group of friends or pack.

It's a great example, with the use of pictures, the simple click, record and send. In many ways not all that different to the voice message feature in Skype. YackPack certainly illustrates where Skype could go with audio / video blasts. There's no reason this couldn't be developed and copied in Skype particularly now that SkypeWeb exists. Is it a priority at Skype? Probably not. Should it be? It's an example of a feature that is "new" thus Skype had the opportunity to innovate in this space and hasn't. When video blasts are restricted to your buddylist "spam" is not a problem. Similarly, additional groups that you might subscribe your SkypeID to would be a value added service. Note that networking groups would probably pay the subscription. Enough said.

Tag : etel

My parents are kleptomaniacs. Just don’t tell them I told you. The garden shed is bursting with stuff. The loft is full of old boxes. The shelves teem with ornaments. (eBay will have a good fiscal quarter the sad day they shuffle off this mortal coil.) And the drawers under the bed are stuffed with toys from our childhood.

Which turns out to be quite useful when you yourself have kids and an endless supply of goodies starts to emerge for free from Nana and Grandad. I’ve been reading this picture book to my older daughter, where a little boy hunts around for his lost pet froggie.

Very cute.

Speaking of which, I think we all know of a very cute voice application that’s currently hiding behind a log and looking a bit lost. Whatever happened to Skype’s mojo? Why wasn’t Skype 2.0’s arrival a case for dancing in the streets?

I can forgive the ringing noise being replaced by an extract from the opera Ode to a Kathmandu Stomach Upset. (Believe me, the full work is quite an experience. I’ve been there.) As a customer, I’m not too fussed whether video is a plug-in or comes out of the box. Tweaks in colour schemes and icons don’t impress me. (I’m male. It’s the way we are. I think my mum is still hoping I’ll notice when they’ve redecorated without having to prompt me first.)

No, what’s important is this. Make it work. And make it easy.

Let’s take the former one first. I bought a Plantronics DSP-400 USB headset a while back. It came “Skype certified” together with a small SkypeOut credit. I’m still happy with it. But it’s also very annoying to use. Because I like to listen to music from my laptop with real, quality headphones. Sometimes I unplug the headset when I move my laptop about, or want to use it on another PC, and Windows takes note and resets my audio devices to point to the built-in stuff. No matter how often I set my preferences in Skype to “Plantronics headset”, it keeps being turned back to Windows default each time I unplug. This is, needless to say (but I’ll say it anyway), not a good experience.

It’s a design snafu in Skype. The actual requirement isn’t to select one device from the list, but I should be able to select the preferred order for audio devices it has seen. If the headset is plugged in, use that. Otherwise, revert to the next preference.

My laptop doesn’t have a built-in microphone, and I never plug one into the microphone jack. I should be able to de-select that as an input. Which means anyone calling me when I have no headset plugged in needs to be warned “Martin doesn’t have a microphone plugged in yet”. This case should be handled intelligently.

Perhaps USB headset owners are too small a minority to make this a priority. But then don’t waste your brand karma on declaring it Skype approved.

Another example is the new video chat. I’m pretty convinced that the video degrades the audio quality. My in-laws have low-speed DSL (256k down/128k up). Just audio is fine; add in the video and the sound gets choppy. This is bad, bad, bad. I can guess what’s happening. Skype uses an audio codec supplied by GIPS. The video is probably a separate thing, and not integrated. Asking GIPS to create an integrated codec that prioritises voice to maintain a minimum bit rate will cost Skype Inc mucho dolula. The audio codec API probably doesn’t make it easy to manage the bandwidth use. So we get a kludge, and the “integrated” product isn’t any better than the third-party plugs-ins. It’s more “overlayed” than “integrated”.

Sadly, Skype’s also having a mid-life attack of feeping creaturism. In the rush to add more feature tick-box items, usability and simplicity is being sacrificed. I think I’d find grouping of contact quite useful. But the user interface is just too ugly and obscure. Hint to Skype UI designers: screens are 2-dimensional. There are information architectures other than “list”! (As I’ve whined on previous occasions.)

And why have my SkypeOut contacts suddenly been elevated into the middle of my Skype contacts? This is a case of the product turning against me. Every time you call someone more than once it asks if you want to name and remember that number. Was a useful and harmless feature as long as the list of remembered numbers kept out of sight. But believe me, the UK Passport Office isn’t one of my buddies. And I’ve only just finished successfully suing Expedia in the small claims court after a car rental screw-up last summer. They’re not my friends, either.

I can only imagine the stress and turmoil the development team are going through as the eBay deal sinks in. But these problems smack of leadership issues, where priorities are not clearly being spelled out, and strategic alignment is being lost. If I were in charge right now, the edict would be “no new features!”. Let the API do the work for you. Focus on making what exists even easier. Make every use case or problem be handled ever smarter. The only exception to the rule is anything that integrates eBay merchants and transactions into telephony.

We’ve already got plenty of VoIP tools that kinda work, sort of, as long as you don’t hit any snags or unusual situations. Skype’s positioning is around simplicity, reliability and ease of use. Lose sight of that, and your cute frog will be forever lost.

Martin reads to us from his Telepocalypse weblog.

Would you like to animate your Skype profile? Reallusion has launched CrazyTalk for Skype which enables messaging with facial expressions and character animation. Thus they say making your Skype conversation more amusing and creative. Your friends will have to get CrazyTalk for Skype too to play. It's a free download to get started.

CrazyTalk for Skype is a dynamic animated messaging tool featuring customizable emotive facial animation allowing you to create Skype characters from any photo! It is much more fun than conventional video chatting. Animate Skype with loads of content you can download from the web or create yourself with CrazyTalk Avatar Editor. Share them with your friends whenever you like, using whatever photo you like, animated however you like – to become the face of whoever or whatever you want to be! Reallusion

Overall, this is another smart piece of piggyback marketing on Skype. It creates new value for Skypers interested in having some fun while introducing the "Skype" community to the oppportunities and benefits in animation. To create your own photo animations you will have to trial or purchase CrazyTalk Messenger. More than a few Skypers should know how to use it and get others laughing. Reallusion is offering a "free" CrazyTalk Messenger so you can animate your own photo's in a promotion they are running. Share with your friends and win.

I know some of my readers will ask why would you do this? I really don't have the time. However I can see an animated character taking a voice mail, or providing some other form of info as just a beginning. For the most part the solutions will just be more fun than that. When Skype adds video (real soon) then even more opportunities for sending animated movies may appear.

Reallusion has effectively created an avatar-like program for Skype that goes beyond what's offered on MSN and Yahoo. I'm sure their technology could be worked into anyone of these messenger programs and perhaps that is what they hope to achieve. For the moment Skype's API wins again in leading the way.

Help me talk better.

There is no way I'm going to write a 50,000 word novel in 30 days. It's crazy. Nevertheless, National Novel Writing Month is in November this year. Hundreds of "winners" cross the finish line through the power of a clear goal (50k words), lowered expectations (this is a rough draft; quality follows quantity), and a deadline.

Many of my conversations would benefit from a deadline.

1. I'd waste less time.
2. Through an ounce of planning, everyone would get what they need from the chat or call.
3. The conversation would be less likely to spill over into the rest of my schedule.
4. It sets expectations for conversation style: short, pointed, transactional, focused.

Two kinds of deadlines:

1. The call. We promise to start this call by 11:01 and end by 11:14.
2. The agenda. Time boxes for talking points.
   • Review/change topics for this call (11:01-11:02)
   • Check in on health, happiness, social lives (11:02-11:04)
   • College update (11:04-11:07)
   • How's the family (11:07-11:10)
   • Send money (11:10-11:13)
   • Schedule next call (11:13-11:14)

This is standard stuff on running better meetings.

I want Skype to help by being more aware of time. For example:

• Launch conversations (voice and chat) from a calendar automatically.
• Remind me of my call/meeting schedule
• Offer to help set up an agenda for the meeting.
• Ping everyone in a conversation with a beep and a private text message about pending deadlines (this topic called "treasurer's report" ends in 1 minute, next topic: "membership report")
• Let us change/revise the agenda in mid-call
• Show a countdown clock in the conversation window with both the big countdown (end of call) and the smaller one (end of topic)

I can start an egg timer or download a software timer. But those are both out of context and not part of the collaboration. Time boxing within the user interface, preserving the visual and cognitive framework of the call/chat will improve the success of the conversation.

Help users and developers build this

This is exactly the kind of value-add I'd like users and developers to build. However today's license, terms of service, and API are hostile to UI changes.

You can see that Skype's design has been amazing about getting users into a conversation, and the hard work of keeping the technical quality high. Now it's time to go inside the call: Help our many styles of conversation be more effective.

I don't expect Skype to help me organize a party, plan a wedding, play a game, hold a quality circle meeting, answer a bomb threat, or talk about my car with a potential bidder. I do want Skype (sometime in 2006, please) to let me use, create and share "conversation helpers" the way I use, create and share templates in PowerPoint and Word. Let the power of millions of users shape conversation to their ends.

Have better conversations with Skype

As with PowerPoint templates, most conversations guides will be free and a few worth money. Please don't think of this as a ringtone opportunity. Think of this as (a) part of Skype's platforming strategy, (b) making Skype more social (as we share conversation helpers), (c) making Skype conversations more productive than conversations in other media.

So often you just reach out and touch someone, a personal connection. But then...

How much do you talk on purpose?

p.s. I'm enjoying No Plot? No Problem! right now. Tips on prepping for and surviving your four week novel writing.

p.p.s Congrats to Hyland Baron for joining the NaNoWriMo team. Hyland makes projects more fun and effective.

I'm really unhappy with the information architectures we adopt to display presence information. Many of you will be familiar with the work of Edward Tufte and his innovative displays of multidimensional and fluid data on 2-dimensional static paper.

We need to do better with presence data, because that data is going be become a lot richer. So, inspired by Tufte, let's see what we can do. I'd like to introduce to you my little pet Tod the Tadpole. As you'll see, I was diagnosed with disgraphia horiffica and have the drawing age of a 3 year old. Never mind.

(A friendly wag suggested this should be Simon the Sperm, but as a family blog I'll demur…)

What this does is adds some more dimensions to our presence display. The most obvious one is a temporal history of our availability. In the example, when the tadpole tail is high, you're available, when it's low you're not. The time scale is squeezed up as a log scale; the last minute and last hour might have the same pixel-width; the far end of the tail might be summarising whether you where around at all last week in just a few pixels.

This history is useful. If someone has just got back from vacation, you want to see that. If someone's online all the time, there's no rush to grab them; conversely, if they're rarely online and you see them come on, call them now!

The up-down movement of the tail is smoothed by adding some inertia; coming online doesn't make it zoom straight to the top, but applies a point force that accelerates it in that direction. (I guess some user testing would tell us whether "y", "dy/dx", or "d2y/dx2" is the right vertical scale.)

Day and night are shown too. This is important when buddies might be spread around the world and very mobile. I've drawn it really badly, but twilight and dawn would be light grey, whereas the middle of the night would be a jet black background. Tod is approaching sleepy time. Naturally, the lengths of the day and night phases would reflect your actual daylight at your current latitude.

You might even choose to colour the daytimes with weather-related information from the locale of the person, such as temperature hues or a pattern of raindrops.

The tail might also encode data about the nature of the presence, beyond being online or offline. For example, the red parts could indicate "busy" — i.e. typing, talking or dragging. (Just clicking in a browser might be regarded as the equivalent of being idle!).

Episodes of mobility, where such data is available from the user device or a network operator, could also be displayed, such as by using a dotted line.

Looking into the future, the background might indicate someone's predicted presence status. If their calendar has a meeting shown, add a border for "busy". If they're due to catch a plane, add a border for "away".

The "head" of the tadpole is also presence-enriched. If you're listening to music, a set of headphone appear on your head. Hey, the little sound-marks coming out of your ears could even beat to the music! Roll the mouse over, see what they're listening to. If you're on a phone call, it looks like you're wearing a headset. And so on.

Of course, the head icons would be personalisable for more immediate recognition. After all, they're your avatar. A great service would be one where you could feed in a normal digital photo of yourself, and it would do all the pattern and colour recognition to churn out a race, age and facially structure look-alike (assuming that's what you want!)

Facial expression could also come into it. A huge chunk of our brain is given over to watching faces, and it's not used much in today's presence and telephony. Don't show a clock icon when someone is away — make them look like they're dozing!

Ideally the head would have a contextually appropriate background, such as a stylised version of "home", "office", "car", "out and about" and "abroad". Tricky with a small icon, but possible if you allow a little more screen real estate.

Which brings me to my last point. Take a look at this miniaturised screenshot of Skype.

Yes, it's weeny. This protects the privacy of those careless enough to become my buddies. But more importantly, it lets you see the overall structure rather than the detail.

What do you see? A ton of white space! Is this vertical scrolled list the best possible information architecture for presence data? I think not. Now, I'm not sure what the right one is. You need predictability of location so you can find folks. You need to properly group and sort according to current presence status as well as tribal affiliation (different work, family and friend groups). There's a lot of variables, and an unconstrained space on which to display them. Other people get PhDs doing this stuff.

Why is better presentation of presence data so important? Because the toughest part of a phone call is the rendezvous. We often miss each other, play phone tag, have hurried "can I call you backs" (and don't). We often simply don't make some social calls for fear of calling at a bad time, and eventually relationships with old friends dissolve. Anyone who thinks telephony is just about creating a duplex audio stream isn't looking at the whole problem.

Anyhow, I eagerly await for someone to rise to the challenge. In the meantime, remember Tod the Tadpole next time you accidentally call someone at 4am — who isn't there anyway.

via Telepocolypse