How AI and VR will enhance language learning beyond 2023
Since around 2016, the fourth industrial revolution has meant colleges, professors and educational practitioners have felt the pressure to integrate advanced modern technology into their classes and keep up with developments. Then, a curveball: In November 2022, the startup OpenAI launched ChatGPT, generative artificial intelligence, showing promise as an effective teaching tool of the future.
As a disclaimer, although language processing models have been impressive and educators are starting to overhaul and adapt classes, these technologies are still imperfect. ChatGPT occasionally spits out misleading or just outright incorrect answers, making some educators crack down on access. Meanwhile, educational institutions still view virtual reality as an overhyped luxury that many can’t afford.
But their potential as educational tools, particularly for language learning, outweigh the risks of trialing them out. Without a doubt, a huge shift in language teaching and learning is incoming, and educators must begin to rethink policies, curriculums, and ways of teaching.
Here is a breakdown of three specific technologies educators should watch out for and how each enhances language learning differently.
1. Cutting-edge AI chatbots
With the lack of in-person practice being one of the setbacks with language learning, autoregressive deep learning models could step in to produce human-like interactions. In the study “Artificial Intelligence (AI) Chatbot as Language Learning Medium: An inquiry,” the results indicated that chatbots have a high potential to be used as a language learning educational tool.
Several free chatbots are currently used worldwide as chatting partners for speaking foreign languages, but they often have word-count restrictions or a paid subscription model running alongside. Some of these tools are powered by OpenAI’s Text-Davinci-003 model, for example, with similar capabilities to the new, advanced language model GPT-3.5-Turbo. It allows students to practice written conversations with different personalities and in various languages.
Reddit has been the playground for hearing users’ feedback, and the reaction is positively overwhelming. Some Spanish language students have said they speak to chatbots every day, as it is a stress-free way to practice.
Chatbots can also be used anytime and anywhere (while it is harder to find regular human chatting partners); they are willing to repeat repetitive material with students over and over again, and learners report feeling more confident using chatbots than dealing directly with human tutors.
One of the setbacks is that the AI models can lie to students confidently and may incorrectly correct grammar. Therefore, the tool is still not paired with private classes or used as part of university curricula. However, it can help fearful students instigate conversations and have regular language practice.
2. ML algorithms for pronunciation
Improving pronunciation has been getting on language speakers’ nerves since the age of time. Even those who can almost speak a language fluently still often have remnants left over from their mother tongue, which can lead some students to resent the language they are learning.
But there’s a solution: Automatic speech recognition algorithms trained by native speakers for computer-assisted pronunciation training. By uploading videos saying difficult words in a second language, novel deep learning methods — such as those used at Amazon — can listen to the speech and give immediate feedback about pronunciation errors.
Although the algorithms produce fairly solid results, with the likes of ELSA Speak leading the way, the accuracy of the machine learning models for detecting pronunciation errors is still affected by the limited amount of training data available.
Therefore, teachers shouldn’t adopt these tools as part of official curricula at schools or universities quite yet. But as they are likely to be ubiquitous in the future, educators and students must at least get familiar with the technology and ask themselves: How can we use this ethically and safely?
3. Virtual reality to transport students to language exchanges
Speaking multiple languages improves employability and gives young people direct access to cultures, arts, literature and content (like music and cinema) that would otherwise be inaccessible.
But the age-old tale is that nothing can replace practicing a new language like being immersed in the country or culture itself. The issue is that these travel opportunities are not evenly accessible to students from different socioeconomic backgrounds.
Virtual reality could change the whole ball game. What if students could improve their fluency in a real-life scenario at a low cost without feeling self-conscious in front of native speakers?
VR can transport language students to a virtual, 360-degree scenario available on mobile, desktop or with a headset. In a couple of minutes, a student could go from their day-to-day routine to buying a baguette in France or sharing a traditional ajiaco (chicken soup) with a family in Colombia.
Alternatively, educators could set up treasure hunts that students can access on their phones. By placing virtual objects in real-world locations, educators can lead students from clue to clue while teaching them about landmarks and history.
If more free apps and software with fewer limitations appear, VR will close language barriers, making language learning more inclusive and minimizing the expectations of spending significant time in another country to learn a language.
The main challenge of learning a new language is often the lack of opportunities to practice. Now, with the latest technologies, educators can support students by showing them tools for conversation practice, pronunciation training and immersion experiences so every student can unlock new language skills independent of socioeconomic status.
Source: https://corp.smartbrief.com/original/2023/04/edtech-language-learning
