In my last post regarding the installation of Project54 on a marine patrol boat, I mentioned that the boat environment presents a new set of challenges. One of these challenges is that it is very difficult to perform manual interaction while the boat is moving due to the rough motion resulting from hitting waves. When officers have to travel faster, for example when they respond to an emergency call, the motion of the boat is very jerky. The difficulty in performing manual interaction in this particular scenario is illustrated in the video clip below. In it, we are traveling at roughly 30 mph and I am attempting to interact with the Project54 software.

In this situation speech will likely be the best mode of interaction, however, as you can tell from the video there is a significant amount of engine noise that can be heard in the cabin. The engine noise can decrease the accuracy of the speech recognizer, and currently I am looking for an appropriate noise canceling microphone as a solution to this problem.

Brad Bell

During the Spring 2008 semester I’ll be teaching a graduate course (900 level) on discrete time speech signal processing. This course will cover topics on the theory and application of discrete time speech signal processing. First we’ll discuss the speech production mechanism and anatomy, as well as linear models of speech production. Next, we’ll delve into speech signal analysis and synthesis methods: pole-zero speech models, homomorphic processing, short-time Fourier transform processing, filter banks and sinusoidal synthesis and analysis. We’ll continue with a look at frequency domain pitch and voicing estimation. These topics are of particular interest to me these days, because one of my PhD students, Alex Shyrokov, is using prosody in modelling human-human spoken dialogues, and another PhD student, Zeljko Medenica, may use prosody in detecting frustration in drivers.

The course will conclude with a look at applications of speech processing, with topics selected from the following areas: speech modification, speech coding, speech enhancement and speaker recognition. Note that some very prominent applications of speech processing, such as speech recognition and text-to-speech converstion will not be covered in this course. These topics would require deeper coverage of statistical discrete-time signal processing.

I will use a book by Thomas Quatieri, called Discrete-Time Speech Signal Processing. The book has many Matlab problems, which will be very helpful. It also has a similar feel to my favorite DSP book, Oppenheim and Schafer’s Discrete Time Signal Processing. I’ve used the Oppenheim and Schafer book as the source of many notes for the Introduction to DSP course (ECE 714/814) at UHN that I’ve taught several times.

I’m looking forward to teaching this course. Feel free to email me if you have questions or comments (andrew dot kun at unh dot edu).

Andrew Kun

Alex Shyrokov was awarded one of three bets presentation awards at the NECHFES student conference. He received the award sponsored by Charles River Analytics. Here’s Alex showing off the award:

Congratulations Alex! 

Andrew Kun

Hi all,

As you know couple of us went to the NECHFES conference in Boston on Friday 11/09. The conference was about human factors and ergonomics. It was my first conference and I was pretty excited specially because I was one of the speakers. We had 10 minutes for the presentation and 5 for Q&A. I talked about my work, Human Computer Interaction while driving.

The conference was very interesting, a lot of presentations were related to driving and driving safety. There were also companies from the filed of human factors which presented their work. We were lucky enough to listen to Dr. Richard Pew, a human factor specialist with more then 30 years of experience in the field. As icing on the cake Alex won the best student presentation award.

Here is a short video of my introduction for the presentation.

Nemanja Memarovic

Hello all,

For some time now, beside other things, I have been working with pressure sensors on the steering wheel. I mounted four of them to the perimeter of the wheel. They are connected to an amplifier board. The signal is then fed to a Texas Instruments eZ430-RF2500 board, which serves as an AD converter and a wireless transmitter. Very interesting thing if you are amazed by small electronic devices, like me. The following figure shows the setup:

The data is collected and recorded at a remote computer. I combined the data from the sensor with the data from the simulator, and visualized it in Matlab. It is shown in the following video. The wheel is displayed in yellow where no pressure is detected. It gradually turns to gray and black depending on the amount and location of pressure. First, the wheel is squeezed multiple times in a clock-wise manner without driving. Second, a city scenario is driven, while the hands are fixed at around 10 and 2 o’clock. It can be noticed that the brightness is changing depending on the amount of pressure on a certain segment.

Now, that the sensor system is completed, it can be used in a number of ways: to trigger a Push-To-Talk signal or to collect tapping patterns and map different taps to certain commands. For example two taps on the left-upper sensor could activate the left index light or lights and siren, etc. It has may possibilities. If you have a good idea, what else it can be used for, please let me know by leaving a comment. Thanks.

On another topic, I recently rediscovered Matlab’s possibility to record .avi videos. It can be a handy tool if you want to present results over time and with more than one space dimension. Just type ‘avifile’ in Matlab Help, the usage is self explanatory from the example.

Cheers,

Oszkar Palinko

Amazon just came out with a “wireless reading device” called Kindle. Kindle has an electronic paper display and this should make reading easier on the eyes than say reading from an LCD screen. You can buy books and have them delivered wirelessly. For this Kindle uses Sprint EVDO. Pretty nifty!

The Amazon Kindle page has a nice video demonstration (scroll down past the product overview). The one thing that worried me at first was that the video implies that, if you want to read your own Word or PDF documents, you need to pay Amazon to convert them to a format Kindle can read, and then email the converted document directly to your Kindle. As it turns out, Kindle can read unprotected Mobipocket files and you can create these with a free tool from Mobipocket. You can then move these files to Kindle from your computer via a USB connection. So, while you certainly can just email the files to your Kindle (or someone else’s) you can just take care of the conversion and transfer yourself. Phew.

Some of the nice features? You can connect to Wikipedia, look up words in a dictionary as you’re reading, get newspapers and magazines and listen to music. A drawback would be that the display is not color, but for reading books and newspapers that’s not a problem. Magazines and technical papers may be another issue.

I quicky browsed the Kindle book selection (Amazon says they have over 80,000 titles available) and saw books on MS Visual Studio, Jared Diamond’s “Collapse,” but not my favorite C++ book “On to C++” (I guess I really need to learn C#).

Anyway, this is a brand new product, so we’ll see how well it does. It’s so new that when I looked it up online this CNET review was about 38 minutes old! I have to say I’m curious how this latest ubicomp device does. At $400 it’s not too expensive, but as the CNET review points out, for that much, you can buy an ASUS PC - read more about that on Andras Fekete’s blog here.

Andrew Kun

One of my graduate students, Mohammad Salah, has recently been appointed Assistant Professor in Mechatronics at the Hashemite University in Jordan. Mohammad worked with me on imitation learning and he got his MS in EE at UNH in 2003. He then went to Clemson University where he received a PhD this year. This is really great news - congratulations Mohammad.

Andrew Kun

The keynote speaker at the 2007 NECHFES Student Conference was Dr. Richard Pew, who is a Principal Scientist at BBN. The speech touched on highlights from Dr. Pew’s fascinating career, including for example his work on the user interface of the Kurzweil 250, the first synthesizer to produce lifelike imitations the sounds of real instruments. Here’s a picture of Dr. Pew with a slide showing him testing out a device for directional amplification of sound which basically tried to super-size your ears! I don’t think he liked it much.

However, in this post I wanted to discuss the concluding remarks from Dr. Pew’s keynote speech. These addressed the question of what the future holds for human factors engineering. First, Dr. Pew expects that the advent of ubiquitous computing will mean that human-computer interaction problems will turn into human-system interaction problems. This prediction is very relevant to the work going on at UNH’s Project54, since we are in essence building ubiquitous systems for law enforcement. I also think that this expectation meshes well with the seamless interaction idea favored by most ubicomp researchers. (Some argue that “seams” are a good thing and are needed to either remind us of the fact that Big Brother may be watching, or to just keep us actively involved in our world.)

Second, Dr. Pew feels that Web 2.0 and mashups will allow human factors engineers to extend the design process into the period of actual usage of a new system. With the new face of the web, engineers can build user (system?) interfaces, deploy them, let users interact with them, and then change the design based on feedback from users. This makes it easier to design good systems than say relying primarily on lab experiments, or even ethnographic studies, which still must be limited in scope.

Third, Dr. Pew expects that blogs, podcasts, and the like will provide a plethora of ethnographic data to support design. So, when studying someone’s work environment, you may not need to go out and join them for a day or week, you may be able to look for data on the web, since the data are already reported in multiple formats. It’s worth mentioning that people certainly do this sort of work with historical data, see for examlpe my post on work going on at Tilburg University.

Finally, fourth, Dr. Pew expects human systems engineering to become a mainstream element of systems engineering. The idea here is that you don’t want to design a system without taking into account the fact that humans will have to use the system! The US military for example is very keen on making sure humans are taken into account when designing its systems. Thus, the Naval Postgraduate School has a program in human systems integration.

For me, Dr. Pew’s keynote was certainly the highlight of the conference, and these four predictions were the highlight of his talk. They point to the exciting opportunities, in both science and technology, that lie ahead for all of us.

Andrew Kun

Azentek has developed a powerful full-featured in-vehicle computer with a built in 6.5” touch-screen that replaces the traditional double DIN sized radio. Highlights include: Core Duo 1.83 GHz processor, Vista/XP, up to 2GB RAM, Bluetooth, GPS, WiFi, Dolby 5.1 surround, A/V IO and noise+echo cancelling microphones for hands free calling and voice activated navigation. Additionally this computer supports advanced telematics applications such as remote fleet management and vehicle diagnostics.

here is a video of the system:

Ian Cassias

del.icio.us (pronounced delicious) is a web service that allows you to store your favorite links online. In addition, you can tag your links so that they’re easy to search and easy to share. You can create networks with friends and colleagues. You can also explore what other people with similar interests are tagging (a cool conference maybe?). 

You’ve probably seen tag clouds (e.g. on Flickr). Tag clouds help visualize what is popular on a site. del.icio.us also allows you to create your own tag cloud, which is a nice thing to display on your home page or blog. As an example, here’s my del.icio.us tag cloud:

Feel free click on “Add me to your network,” perhaps we can exchange some interesting links!

Andrew Kun