But for some reason I just did not relax. I didn’t get to spend as much time with friends or just vegging on the beach as I wanted, and I really have no one to blame but myself. Part of the problem is that I just could not shut my brain off. Last year I got a notebook, and it ended up being my ‘PyCon’ notebook. This year, I just didn’t seem to have time, oddly enough. Code freeze at work was Friday and that, I am sure, didn’t help. The official fork will most likely be this upcoming Friday.

I have had no time to work on any of my python projects, and it is driving me crazy. There are so many fantastic things happening with django-survey, and Pinax is picking up steam, I need to put the old PyCon ‘08 stuff into archival mode, and start up ‘09. There is some very very interesting stuff going on with the DFW Python group that I want to help out on as well. There are so many fantastic things being worked on right now by incredible people, and I feel a down right claustrophobic not being able to do anything myself. I only been able to attend one Boston Python Meetup so far this year!

It looks like I will be in Montanna the first week of July this year (my anual pre-PyCon-Tech kickoff-kickoff). Barring any project ‘issues‘ I hope to get my act together then (with respect to PyCon software for 2009). If there are any pythonistas in the Missoula area, please send me an e-mail! For now I guess I should get back to unpacking the car.

I saw some of the most fantastic talks today. I could talk about those, but that is not what I will do.

I was stressing out today as I had a lightning talk to give. Lightning talks are nothing in general, except mine was on speech recognition and why Nuance is being a sponsor. I love speech recognition, but doing live demo’s of it are next to impossible. I ‘have’ the equipment to do it right, but everything fell apart during the past two days. My very expensive noise canceling microphone headset which can deal with audio feedback from monitors is back in Boston. The laptop that I was supposed to use never showed up. The USB stick that my mobile profile and slides were on was destroyed by the swag. The install CD’s I had were destroyed at some point; thanks to a spilled coke I think. The headset alone is a deal breaker actually. A normal human being would have given up at this point, but as anyone will tell you, I am anything but ‘normal.

I did have a hard drive with an early gold candidate of Dragon Naturally Speaking 9.5 (totally by accedent), Catherine Devlin graciously allowed me to use her laptop, and Carl Karston and the AV crew <insert forgotten name here> set me up with a Sure USB Microphone. This Microphone was the shiznitz! So I scrapped the entire planned talk and decided to work with what I had. I think it worked out in the end.

I should explain a bit about doing live demonstrations of speech recognition. They work quite well in demonstration booths if you have an enrolled profile (the user has trained with the system), and a high quality $$$ mic. But there are some rules to doing these demonstrations. You do not do them with a bad microphone. You do not do them on unknown laptop hardware. You do not do them with a speaker independent un-enrolled profile. You do not do them with out a prepared script that you have tested and know inside out and backwards. And you never, ever do them in front of 1000 people in a very noisy, echoy ballroom external under amplification. If you break these rules, you end up with demonstrations like the one Microsoft had, and it ends up on youtube (out of professional respect I will not link to their demo), and it continues the myth that speech recognition doesn’t work. Because it doesn’t work under those conditions, as even humans can’t recognize what you are saying half the time.

My original talk was going to be fun and interesting and poke fun at speech recognition demonstrations in a way that people would get, yet still show off how good it was even in bad conditions. The software can handle it; I have tested it enough to be sure of this. Well under the actual conditions I was faced with I knew it would be Microsoft all over again. I had four options:

1. Follow the script and make a Nuance products looks like total garbage.
2. Ditch the live demonstration and give a dry boring talk on why Nuance was sponsoring Pycon (and have it forever forgotten)
3. Ditch the talk completely, and leave Nuance as just a footnote on the website.
4. Have fun, and if I get very lucky, people will remember Nuance in a positive light.

I am not 100% sure I got very lucky, but I did get lucky.

Only at PyCon can I show up with litterally nothing, and have people bend over backwards to help me. I just asked Catherine to use her machine because I noted at the python lab that it was a nice, fast Dell XPS machine. I had 0 expectation, and yet she had absolutely no problem giving me her machine and password for full access to it for an in determinant amount of time. Carl and the AV guys were running around like mad looking for anything that would work. What they came up with was beyond my expectation and set the entire tone for the talk. When I saw the mic I had a flash of inspiration, and somehow it seems to have worked. I actually did a 3min enrollment with the mic in the atrium in a deafening din of noise during the break just before the talk, much to the enjoyment of Mary who was almost in tears with laughter. (I was using our David Barry enrollment script and almost yelling it.) I can’t wait to see the video.

Another surprise was that there were many Dragon users in the audience.  Many people came up to me after wards when I was doing the demonstration in the (now quiet) Atrium with success stories and offers of professional headsets. One person (Lisa) is a storm chaser and uses dragon while driving after and away from tornadoes! Other people came up with very good questions. The level of sophistication of the average PyCon goer is quite high and everyone had a great respect for the problems of speech and language in general. There were none of the ’star-trek’ questions I often get.

Later I ran into someone who had had less than stellar experiences. We talked for quite a while about what the source of the problems could be, and I am very pleased that they are willing to try again and provide feedback. For some people speech recognition using the standard speaker independent base models just does not work well. There are so many variables determining why this is the case on a person by person basis, that it is next to impossible to tell what the real problem is just by talking. These users are the ones we want to know about. Those who try it, but for what ever reason it just doesn’t seem to work for them soon enough. Those with only ~95% accuracy or worse and quickly give up (or even not so quickly). Due to the activation/registration that the product does, we have decent numbers on how many people are effected by this. The numbers are quite low and it is not a problem from a ‘marketing’ or ‘business’ perspective. It is a problem from a research and understanding perspective. We want it to work for everyone.

Other important notes:

• playing 4 player packman, glaxon, nibbles, and attack %^#^$% ROCKED! Phil is my hero…
• The twisted folks are the coolest people on the planet. (Being from Boston doesn’t hurt)

My god, I wish I had something good to say about this piece of advertisement. Lets decompose.

1. The product

They somehow took a product that is fun to watch people use, and has a very cool, minimal, embedded interface, and made it a chore to watch. Worse yet, they misrepresented it’s capabilities. It works better, faster, and easier than is portrayed. It is a continuous speech recognizer, why is that kid talking discreetly (one word at a time, and slowly?). It looked like he was using Kurzweil Voice or Dragon Dictate from 1995! Even the executive is talking as if he is reading a cue card (which is most likely the truth of it). This software is used by stenographers in court rooms. The fastest speaker can be recognized. I thought the most precious thing in a commercial was time? Why is there so much filler?

They never actually show the software clearly. The dragon bar is just a yellow blurry blob on the screen. Granted you can hide the bar, but the end result is, the system appears to be very slow and unresponsive. Even the cute typing text in the add is a misrepresentation. We do not send one letter at a time to the screen; that would be annoying after 2 seconds, not to mention slow. At least the womans use of the product is a little closer to reality. Oh and yes, you have full command and control over your PC ala Star Trek, but by saying ‘listen to me’ and ’stop listening’ instead of ‘computer’; though you would never get that impression.

2. The script

Wow. I mean wow. When it comes to emerging trends, technology, and what the ‘hip’ kids are doing, our upper management is fairly on top of things. They also understand the products they sell at a level I have not seen at any other company. I guess the same can not be said of our hired marketing firm (or whoever produced this thing)! The things wrong with the script (from both a technical and substantive perspective):

• The woman writing an e-mail: Even my Mom knows what a signature is and how to set her mail program to include it automatically. Who in their right mind would use a voice macro to do that? There are much better ways to show off macro’s. And form filling would be better for multiple signatures. That is a killer feature!
• The history report. I know adults think teenage boys are stupid and cannot write a decent report, but this poor actor must have been choking back bile when he saw this script.
• The IM interaction. This is just abysmal: ‘Are you in’? Every IM client from hour 0 has had the ‘away’ feature, so yes, he is ‘in’, you know that already. I guess ‘in’ is the hip new lingo all the kids are using these days.
• The executive. This is how a teenage boy pretending to be an executive talks.

3. The acting

All I can say is I really hope we got a good deal here. In some cases I think it is the material. There is not much you can do to make it palatable, so why waste the effort. It is a cut rate commercial for a few basic cable channels after all. What is interesting is we have many commercials for our vendors and resellers, and the acting there is top notch. You can find most of those on youtube, or on our website (though this is a demo, not a commercial).

4. The Direction and Cinematography

Ok, its a commercial. The sets were believable. the lighting spot on, the sound perfect, and the basic camera work was good. But please show the product. You never got a good look at what was going on, on the machine. The product is supposed to be the star of the commercial. Here it appears to be a character off screen being talked about.

What would I have done?

Well I would do a rip off of the HP ‘talking hands’ commercials. This was actually my idea back in 1997 when I first started working with speech. Show the apps swapping around like you do in real life, only by voice. Show the real speed of the product. show the people, but not their heads. Show their feet propped up on the desk in front of them. Put the product first, and make it interesting. Have them standing there, and the transparent screen in front doing all the cool things they do in the HP commercials, only they have their hands behind their back. Show people really using the product, not selling it. I guess that is why I am an engineer and not in marketing.

Additional Notes:

No you do not need to speak punctuation. There is auto-punctuation, but before it is accurate the system needs to have been used for a number of hours to collect enough statistics on how you use punctuation and inflection. Having DNS inspect past e-mails and documents can help, but gathering the inflection data is the key part. For that reason it is turned off be default. More research and data will fix that.

No you do not need to spend hours or even minutes training the system. You can use it right out of the box, and it will get better with time. To be honest the learning curve is more on the user end, than on the software end. Talking to a computer is just like any new means of input, like cell phone SMS tapping, or pen tablet. It takes time to become comfortable with it, but not as much time as other input devices.

This blog post dictated with Dragon Naturally Speaking 9.5 Professional.