624 588 Princess Lanting

EnglishCentral IntelliSpeech℠ Technology

EnglishCentral IntelliSpeech℠ Technology

EnglishCentral’s “special sauce” has always been our internally developed IntelliSpeech℠ assessment technology. Teachers appreciate how IntelliSpeech℠ motivates students to speak outside of the classroom. Students appreciate the instant feedback and the game dynamic of trying to improve their speaking scores.

EnglishCentral’s IntelliSpeech℠ is trained on hundreds of millions of lines of speech from students from over 100 countries, using the latest Machine Learning techniques. It assesses learners’ speaking ability across the following dimensions:

Pronunciation measures the acoustical qualities of students’ speech over 64,000 possible triphones (combinations of phonemes), stress and prosodics. 

Fluency is still based on duration and pause rate of the speech.

Completion is still based on whether the user speaks all words, or drops words.

Error Types

As the learner speaks, Intellispeech℠  provides feedback according to the following types of errors:


Line Score

Learners get a line score between 0 and 100 points for each line spoken. Learners lose points for each error made. The amount of points learners lose per error is weighted based on the number of words in the line. For shorter lines, learners lose more points per error than for longer lines.

Only the final version of each line spoken counts towards the users final score on the Video Grade. So, if a line is repeated several times, only the last version counts towards the Video Grade.

Video Grades

The video grade is the cumulative measure (i.e. the weighted average) of the line scores received by the learner from speaking lines in the video.

Intellispeech℠ computes a percentile relative to other learners to produce a video grade. For instance, if we determine that the speech for the video was 75% better than other learners, the learner would get a “B+” as a video grade.


Percentiles are mapped to grades as follows:

Pronunciation Center

The Pronunciation Center (“Pron Center”) on EnglishCental is where students’ feedback and progress on their pronunciation is tracked and where they can find exercises to help remediate their pronunciation challenges.

At the heart of the Pron Center are the 4 steps of our Pronunciation Learning Cycle:

  1. Speak Anywhere on EC
  2. The Pron Center automatically collects all words students have trouble pronouncing during any speaking activity on EnglishCentral, such as speaking lines in our interactive video player, or speaking words in our Vocab Builder.

  3. Pron Center Collects Weak Words
  4. Students then use the Pron Center to focus on their weak words, either focusing on their weakest words, or their weakest Phonemes.

  5. Practice with Courses & Tutors
  6. To help learn how to form Sounds correctly, students follow one of EnglishCentral’s pronunciation courses, or take a GoLive! lesson, 1-on-1 with a tutor and get feedback on their pronunciation and fluency.

  7. Master the Sounds
  8. The Pron Center tracks student’s progress on each word, and automatically removes the word from study when students successfully speak the word 3 times in a row anywhere in the site.

Pronunciation Courses

IntelliSpeech℠ has analyzed our users English pronunciation over hundreds of millions of recorded utterances and identified the most common problematic sounds of speakers for each native language region. Based on this analysis and speech data, we have designed pronunciation courses — Top 10 Challenges — where learners can focus on the most challenging Sounds for speakers from their native language, using authentic videos. We currently have customized pronunciation courses for: Japanese, Korean, Spanish, Portuguese, Turkish, French, Vietnamese, and Arabic.

pron course

Our Speech Database

For the last 10 years, our learners have provided us a trove of data (over 600,000,000 speech utterances from over 100 countries) on how they learn English.

Our reference models are also training on large amounts of data collected from native speakers speaking the authentic speech from our videos (as opposed to many other corpora which may contain artificial “read speech”).

We use this data to create a “Machine Learning Loop” that combines this large data set with feedback from a team of over 600 trained professional English teachers who have analyzed student speaking ability in over 500,000 1-on-1 live sessions. The result is the “machine” (our online self-study platform) learns from the feedback teachers provide, and at the same time, teachers learn from the feedback the machine provides on learners’ strengths and weaknesses.

400 184 Princess Lanting

Why our classrooms are failing language students


Is this an effective use of class time? 

Let’s be honest.

Language classrooms don’t have a stellar track record when it comes to students achieving proficiency in the target/taught language.  This despite, hard working, well trained and dedicated staff and teachers.

Why is this?

In my over 25 years as an educator and working in and with schools teaching languages – 3 very strong reasons spring to mind.

1The classroom experience does not contain “agency” or purpose.

The classroom remains an isolated practice area with little relevance to students.  Students become de-motivated when language isn’t the real thing, when it isn’t used for a purpose and is just words in a list or  a mental exercise of memory.

This recent post by a teacher trainer, nicely illustrates the point.  The teacher despite designing a stellar lesson is confronted by student apathy. The teacher planned a very creative lesson about eating food and ordering food in a restaurant in Mexico.  When she asks why one student isn’t into the lessson, the student replies, ““Why the hell would someone like me be traveling in Mexico?”

The point is, students get frustrated when they don’t see the relevancy of the classroom materials and activities. “The future” is a long way away and provides little practical motivation to a student.  Students need to interact with examples of real language, participate and communicate with real target language speakers.  We need to make our classrooms NOT have 4 walls.

2The classroom does not offer students enough time on task with the target language.

All teachers know about this pink elephant in the room – time.  There just aren’t enough hours committed to language instruction for students to make adequate progress. Language learning is time intensive.

Students need more time interacting and experiencing the target language, more time the school schedule doesn’t offer.  Further, the classroom is for the most part a very ineffective use of the time given. So much time gets eaten up with non-subject related “stuff”. Announcements, classroom management, entry/exit etc ….  We need to design a better classroom experience for students and use this precious time better.

I began helping build EnglishCentral because I wanted a solution to these two fundamental problems plaguing language education in this day and age. I wanted EnglishCentral (and ed. tech in general) not to replace the classroom teacher but to support them and at the same time mitigate these 2 fundamental problems.

How does EnglishCentral do that? 

Essentially, EnglishCentral allows schools and teachers to “flip” their classroom.  Students get additional practice with real life language.

EnglishCentral provides authentic input amd takes a natural language approach. Either through the best videos on the web or through our 1-1 tutor GoLive! lessons about the video lessons.  Students interact with real world language that is relevant to their interests and curiosity.  We make language learning relevant and about “out there”.

EnglishCentral also provides a study platform where through mobile learning, students can more efficiently use their non-class time to get much more engagement with the English language.  Students using their smartphones, dramatically increase their time interacting with the English language.  EnglishCentral allows them to use our apps when waiting for a bus, when waiting on a friend, in the city, on the street, in the park.   We compliment the in-class learning by providing meaningful, “comprehensible” input for the student. More time on task = more gains in student oral proficiency.

Find out more about the blended learning approach using EnglishCentral.  Also how you can flip your language classroom using EnglishCentral.




400 390 Princess Lanting

EnglishCentral Deploys with Huawei at Mobile Summit


EnglishCentral demonstrated its newest mobile app at the 2015 Innovation and Transformation Summit from October 22 to 23 at the Westin Resort in Bali Indonesia.

During the event, EnglishCentral  launched its HiStudy app, the English learning app developed with Huawei for the Chinese Market.  This application has been a huge success in the Chinese market, reaching over 10,000 activations per day, reaching over 10 million in its 1st year.

Huawei is the largest mobile handset provider in China and #3 worldwide, and expanding rapidly in international markets.  We are excited to duplicate the results of the Huawei-EnglishCentral China deployment with mobile carriers and operators throughout the rest of Asia.

EnglishCentral’s analysis of data from over 150,000 learners learners of its mobile app in 2015 shows that students make progress two times faster when they are able to practice from their mobile devices.

For more information about EnglishCentral’s participation at the Huawei Summit, or if you would like to see a demo of the mobile app, please contact us through this form.
[vimeo 140929004 w=500 h=281]

225 225 Princess Lanting

Evaluating The Impact Of EnglishCentral: New Research


EnglishCentral is constantly being evaluated by professionals in the field.  One recent study is by Shane Dixon of Arizona State University titled:”Evaluating the Impact of an Online English Language Tool’s Ability to Improve Users’ Speaking Proficiency under Learner- and Shared-control Conditions

Here is a brief summary of the research and findings.  

Download the full report

The study looked at EnglishCentral as used by 83 advanced level students in ASU’s AECP (American English and Culture Program) listening and speaking course.  All learners were given a pre and post test to measure their English language fluency using the Pearson Versant Test

There were 3 groups examined. 

1. Learner control.  EnglishCentral “free” study. Students chose the videos lessons or courses they would study on EnglishCentral

2. Shared control.   EnglishCentral controlled study. Experienced teachers selected video lessons and assigned them as a custom course for student study. Additionally, students could also select their own video lessons.

3. No treatment.  No EnglishCentral study but were given an equivalent amount of traditional homework as the 2 other control groups

Research Questions.

1.  In addition to the 168 hours of classroom instruction does the use of EnglishCentral (learner or shared) lead to gains in fluency vs the no treatment group?

The study found significant gains in fluency through the use of EnglishCentral as a study tool vs the no treatment group when in the shared control condition.

2. Is the shared-control or learner-control system in the EnglishCentral environment better at achieving learner gains in speaking proficiency?

Surprisingly, the shared-control group (which controlled for language level/appropriateness) had the more significant gains in speaking proficiency.

3. Is student attitude, operationalized as the combination of motivation, ease of use, and feelings about technology, affected by the learner control and shared control models? Do other variables such as age, gender, first language, and teacher effect learning outcomes?

Student attitude and motivation was a factor. Students with higher comfort using technology generally had higher achievement scores. Teacher attitude towards technology and EnglishCentral use in general was not a contributing factor towards student success (as measured by an additional survey).

The research raises many interesting questions which are described in the full paper.  

Is EnglishCentral better used during class time (as opposed to homework done outside of class, in the study)?

Why is EnglishCentral not as effective when students are given full control of the learning environment?

Read and find more research regarding EnglishCentral here and here.


236 267 Princess Lanting

Giving Students More Control: Learning 3.0

networkI call what Alan November advocates, “Learning 3.0”. (see the video below)

It is about giving students the space and tools to make their learning relevant to the world. REALLY relevant and actual. Make them active in the community, bring them alive to their potential.

THIS is schooling, learning, education and whatever sticker you want to place on it. Authentic material. Real problems = real learning. Purpose. Learning for the world, not for teachers or just the piece of paper or to get through the year. Knocking down the 4 walls of the classroom.

It is so important that we get off the assembly line and start really letting students venture/learn in and of their own motivation and volition.

This article – “How Teachers & Tech Can Let Students Take Control“, expounds more on the points Alan November makes in the video.  Technology is allowing schools and teachers to treat students less like children, to really make them into the self-correcting, autonomous, independent learners our society needs and our future demands.

We at EnglishCentral see what we’ve built as being central to this change in education. Making school less about babysitting students and more about giving them additional responsibility for their learning – letting them be the ones driving the “learning” bus.

Just an ordinary talk with an extraordinary meaning to those who really want to change things….  Students can add value to the world. It’s up to us teachers, administrators, publishers to let them have a go.