Great white paper here, particularly the categorization and listing of software and services. Also, good to see an extended discussion of what an LMS is and what one can expect from it.
Do I find this implementation racist and insulting to non-Caucasian teachers, hell yes! They are taking a teacher from the Philippines and putting a white women’s face on them. This could have been done a lot more realistically with true videoconferencing without having to add a layer of animation, which will end up deadening much of the facial cues leaving the animation nearly useless in this regard.
One defense of this could be that they want to maintain consistency in the appearance of the teacher even when using a variety of staff on the back-end. However, that still doesn’t explain the white-washing.
With that said, this is the first implementation of robot teachers (robot-assisted language learning–RALL) I’ve heard about that make some sense. This is basically videoconferencing with a mobile monitor (with some ability for arm and leg movements). This is much better than the stuff presented before that is basically a rolling tape player with some ridiculous voice recognition. Having a skilled teacher behind the robot is key and, at least for the foreseeable future, the only way to provide an optimal educational experience.
As I saw in a presentation by KICE a few months ago, the biggest problem with using videoconfernced teachers in classrooms is having a trained (and motivated) teacher/facilitator physically in the classroom with the videoconferenced teacher. Local teachers tend to sit back and watch or, worse yet, just leave the classroom during the videoconferencing time. In addition, there is little planning time afforded local teachers, thus they tend not to plan classes with their videoconferenced co-teachers. This finding is nothing new to native English-speaking co-teachers in Korean public schools. The same complaints have been heard for years. They were hired as co-teachers, but end up planning and conducting classes alone as a result of both teacher apathy and poor oversight and training.
With all this in mind, I’d say where is the time and money for training local staff? Without it, these high-tech innovations make for great publicity, but lousy education.
Mispredicted Words, Mispredicted Futures
The accuracy of computer speech recognition flat-lined in 2001, before reaching human levels. The funding plug was pulled, but no funeral, no text-to-speech eulogy followed. Words never meant very much to computers—which made them ten times more error-prone than humans. Humans expected that computer understanding of language would lead to artificially intelligent machines, inevitably and quickly. But the mispredicted words of speech recognition have rewritten that narrative. We just haven’t recognized it yet.
After a long gestation period in academia, speech recognition bore twins in 1982: the suggestively-named Kurzweil Applied Intelligence and sibling rival Dragon Systems. Kurzweil’s software, by age three, could understand all of a thousand words—but only when spoken one painstakingly-articulated word at a time. Two years later, in 1987, the computer’s lexicon reached 20,000 words, entering the realm of human vocabularies which range from 10,000 to 150,000 words. But recognition accuracy was horrific: 90% wrong in 1993. Another two years, however, and the error rate pushed below 50%. More importantly, Dragon Systems unveiled its Naturally Speaking software in 1997 which recognized normal human speech. Years of talking to the computer like a speech therapist seemingly paid off.
However, the core language machinery that crushed sounds into words actually dated to the 1950s and ‘60s and had not changed. Progress mainly came from freakishly faster computers and a burgeoning profusion of digital text.
Great blog post (long) on speech recognition and the lack of progress experienced in recent years. He makes a great argument. However, you must check out the comments as there are many excellent responses that counter his arguments and some responses from him to those responses.
1. TubeReplay
2. DragonTape
4. SynchTube
5. Infinitube
6. Splicd
7. TubeChop
8. YouCube
9. MixTube
10. YouFlow
11. Quietube
12. YouTube TestTube
I use YouTube so much in my classes. It’s nice having some of these tools to work with. TubeChop has been one of my favorites. It let’s you grab just the scene you want and embed it.
Some of these others look quite cool too.
I would love to experience this in person. This is an amazing program. I’ve been saying for years that we need to use technology to connect qualified masses of teachers to those who need them around the world.
This is not a cheap program, but it is a well-integrated program that includes the use/adoption of Korean curricula, (ideally) collaboration between the the teachers, and the provision of native English-speaking teachers to areas of Korea that have trouble staffing them.
I saw a presentation as well that painted a pretty good picture of the program. Unsurprisingly, as I alluded to above, collaboration was the real issue. Korean teachers did not collaborate as much as was expected. This isn’t too surprising as it is the same result that we have seen many times in collaboration studies in the States. Most importantly, the Korean teachers were not compensated (in time or money) for the “extra” collaboration. The American teachers were paid for all of the time that they spend on the service so it’s not too difficult to see why they were willing to do so.
If this is really going to work throughout Korea, much still needs to be done. First, training in collaboration and methods specifically revolving around the use of videoconferencing needs to be added for pre-service programs and in-service training. There is no way that teachers are prepared to do this. These are some of the same problems we see in NEST-NNEST collaborations now in Korea. We are going to make the same mistakes again if we continue to put the technology before the training.
Interaction in Blended EFL Learning:
Principles and Practice.Mei-Ya Liang and Curtis J. Bonk
Taiwan / USA
Abstract
The trend of acquiring English as a foreign language (EFL) through blended learning (BL) has prompted teachers to develop strategic plans and directions for its onsite implementation and evaluation. This paper applies the concept of interaction to the challenge of creating a BL curriculum for an EFL class. General principles of interaction based on three dimensions of interaction—textual, social, and technological interaction—are presented and then applied specifically to EFL classes at a Taiwanese university by adopting the following practical steps: (1) setting course objectives; (2) formulating techniques and strategies; (3) selecting media and tools; (4) organizing activities and technologies; and (5) evaluating student learning. Students’ reactions to and comments on six BL curriculum units indicate that various combinations of BL based on level and dimension of interaction are well adapted to the specific university EFL class. Our findings suggest that the interaction-driven approach should be the focal point for future development and implementation of BL in EFL classes.
I’m crushed for time right now, but I’d like to look further into this. Blended learning seems such a natural mix, but like everything, we need to know what the best ways to design for this approach to teaching and learning.
I have to say that this is nothing really new conceptually. This is essentially the interactions that have been taking place is SecondLife for years. The interface is different though. This is likely using something similar to Playstations video or Microsoft Kinect.
The latter is probably the one that will have the most impact on a project like this. Kinect (with the Xbox) will make interfaces like this much easier for developers to integrate into their software.
From the article, I can’t tell what the role of the guy in the picture is. Does this software require a real life dialog partner? That could be cool for practicing more realistic dialogs between language learners, even without the teacher. Otherwise, I’m wondering what the purpose really is. Having smart bots engage and respond would be much better overall.
Babbel
, the language learning site, has added “realtime” speech recognition to enhance its practical application and enable users to fine-tune their pronunciation skills. This pits the service up against more traditional players such as TellMeMore
This could be really cool. I like babbel and I think that even though speech recognition is not great, it can still add some needed emphasis on pronunciation.