Language Speaking Practice Solutions Meet Artificial Intelligence: A 10-Year Journey With Poodll

283

Presented by Poodll, the best language speaking tools for the Moodle LMS. Get an exclusive $150 USD coupon for Poodll Essentials Plus and enjoy unlimited interactivity — Ask now

Post Pages - Post Inline - WIRIS

Adapted from a presentation at MoodleMoot Colombia 2020 Online.

Two sprawling technologies, each reaching newer heights at faster rates, transforming the lives of more people around the world, yet solutions that remain apart. A powerful coalescence is still yet to be seen.

Between Artificial Intelligence (AI) and elearning technologies, the journey of Justin Hunt and Poodll, an open source recording plugin family that a decade after its creation would exemplify the “EdTechPreneur” success story, whose mass adoption would justify the existence of a product and a business model instead of the all too common other way around. Poodll’s comprehensive set of technologies, tools and recipes for language learning continues to outshine the competition, whose tech fails to capture the nuances of effective speaking practice, by figuring out the precise ways in which AI technologies —Natural Language Processing (NLP) and voice recognition in particular— can actually improve the experience, by adding both new and useful things; as well as filling in the gaps of a smoother path to practice. Because when it comes to language speaking skills, the science is in: Practice, practice, practice!

If you are interested in helping make this relationship not just happen, but matter, Hunt invites you to put on your educator cap as we take a short trip into the recent developments in AI.

Elearning Success Summit

5 AI and elearning avenues. Sinergy, or collision course?

It would be odd to find objections in favor of the use of AI and other “next-gen” technologies to improve the elearning experience, particularly in the areas with assignments due such as speaking practice, whose scalability has always depended, at least partially, on the evolution of reliable voice recognition tech.

Zooming out, the issues of voice and language processing for practical applications reflect a global phenomenon that became more prescient on the second half of the past decade. It is about the place and promise of AI into our everyday lives. The latest update in LMSPulse’s annual “AI Buyer’s Club” guide frames the problem fivefold:

  • The technologies proper. How technologies are advancing, which ground (if any) is being broken, and how the technological value chain is standardizing, in order to guarantee a reliable innovation stream.
  • The economics. Were it not for pesky economists and their material concerns, wild or utopian ideas would be entertained. But as turns out, a series of technical, financial and human boxes need to be tick in order for a technology to deliver valuable and remain with us by the will of the invisible hand alone.
  • The consumer. If it is to be considered part of the economy, it is the “irrational” part. The right brain. Needs met and verifiably quantified belong in the same mesh with spur-of-the-moment trends and ancient traditions. The mesh makes AI look god-given at one turn, and a safe road to ruin at another.
  • The elearning ecosystem. It is assumed that the success of radical technologies relies in their ability to interact with existing technological stocks, hopefully to expand them and maybe ultimately replace them. It is safe to propose the Learning Management System as the epicenter of the stock and the integrations, therefore the success of a given AI in the ecosystem depending on its ability to “play along.”
  • The AI deployments. Apps, or in the case of LMS integrations, plugins or add-ons. Do they understand their environment and end user, while maximizing the value delivered to both?

At the turn of the decade, from recommendation engines, to disease predictors, to autonomous vehicles, it remained to be seen.

2016: The mobile prerogative

What if AI devises universal understanding, or expressions of beauty, that the human mind finds itself unable to comprehend?

Reality has turned out to be less dramatic. Early on, conventional wisdom settled on making the evolution of AI reliant on large datasets —instead of, say, radical new ways to make algorithms “intuitive”— thus a brute force race was born. Whoever collected the most data would be declared champion. Somewhere in the middle of the decade a constitution was drafted by a handful of 20-year old boys, sparking a multi-trillion industry whose structure did not give a lot of people a say.

Data was chosen as the new oil. Coupled with an overdeveloped focus on customer behavior, the focus of the algorithms was put squarely on human information: Personal attributes, personality, demographics, behavior.

You would think these massive collection of personal data would render fantastic insights into how students learn, therefore radically new pedagogies, orders of magnitude more productive.

Well… We’re still hopeful!

Thanks to mobile devices, the increase in volume and variety of data collected has skyrocketed, but it has not led to an equal increase of innovation. But it has made some billionaires.

To be continued…

Elearning Success Summit

Augmenting the language learning teacher with Poodll: Justin Hunt at the Elearning Success Summit.

At the Elearning Success Summit, we got a unique opportunity to listen Hunt talk about his storied career, as both an open technologist and language teacher. As a language teacher abroad, Hunt experienced first hand, how hours and hours, sometimes amounting to years of language practice, just did not translate to language skills.

On the technology side, we’d also realized how the incursion of computers in education —Are they still called Computer Labs?— were led by engineers aiming to replicate the unaided classroom experience rather than “augmenting” it according to the language teacher’s design and evidence for effective practice. “It was great technology,” but for language learning “it was useless,” Hunt recounts.

He set it upon himself to make the most of the technology available to enhance the language learning experience.

After a decade, this simple principle has proven highly effective when it comes to stay relevant and maintain a rate of innovation. Poodll’s pioneering foray into voice recognition makes perfect sense from this vantage point. Today, the variety of tools and apps to listen, reading or writing English are, in Hunt’s view, “pretty good.”

Today we are able to do so much more than we were able to do.”

This article is made possible by Poodll, the most comprehensive solution for speaking practice, language learning, and A\V recording for Moodle. Create assignments and provide feedback through html5 audio or video recording, with instant AI-based transcription and secure storage on the Poodll cloud.

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.