For online, traditional schools and language educators
Fluency EI — Educational Intelligence for Schools
Understand every student's true speaking level.Across languages. Over time.
An independent CEFR benchmarking and progress-tracking layer for schools — not a conversational practice app.
Fluency EI proposes. The teacher decides.
Assessment frameworks
- CEFR — Active
- ACTFL — Coming soon
CEFR is the only active framework today.
No curriculum changes · No new testing framework required
Camille Moreau
Cohort · Intermediate · FR-204
Current CEFR
A2
Trajectory
Progressing → B1
B1
B1
A2
B1
A2+
Key insight
Strong vocabulary range, but recurring errors in past tense (passé composé).
Progress over time
Camille Moreau · 7 assessments
Trajectory
Steady progress within A2 — on track to reach B1 in the coming term.
Trusted by language programs at
Lycée NumériqueÉcole ConnectInstitut PolyglotteAcadémie LumièreLanguage learning is still measured the wrong way.
Most schools rely on static tests and occasional assignments. Performance varies depending on the day, stress, or topic — and a single score rarely reflects a student's real level.
Between assessments, teachers operate with limited visibility. Learning gaps form quietly, and by the time they surface, students are already misplaced or disengaged.
Introducing a new testing framework or restructuring the curriculum to fix this creates friction with educators, parents, and platforms — and is rarely worth the disruption.
4 weeks
suggested interval between assessments
<10%
illustrative real-time visibility between tests
Late
detection of learning gaps
Suggested reference figures, not measured school statistics or required settings. Each school decides how often to assess; one assessment every four weeks is a possible starting point. Schools will be able to adjust their schedule in Fluency EI when the school configuration module becomes available (planned).
A persistent intelligence layer, without changing how you teach.
Fluency EI sits on top of your existing systems and analyses the data your students already produce — no new testing framework, no curriculum redesign.
Student recordings
Speaking exercises and oral submissions are analysed for pronunciation, fluency, and hesitation patterns.
Written assignments
Essays and short responses are evaluated for grammar accuracy, vocabulary range, and complexity.
Test responses
Existing tests and quizzes feed into the model to track trajectory across CEFR levels.
AI Language Level Assessment.
Students continue their normal learning activities. The platform evaluates performance across five dimensions and maps each student dynamically to CEFR levels — A1 through C1 — updated with every new piece of work.
Assessment frameworks
- CEFR — Active
- ACTFL — Coming soon
CEFR is the only active framework today.
Student is currently A2, progressing toward B1.
Strong vocabulary, but weak past tense usage.
Camille Moreau
Cohort · Intermediate · FR-204
Current CEFR
A2
Trajectory
Progressing → B1
B1
B1
A2
B1
A2+
Key insight
Strong vocabulary range, but recurring errors in past tense (passé composé).
Performance Benchmarking Engine.
Fluency EI compares each student's real performance against expected CEFR benchmarks — defining how students at each level should actually perform across multiple linguistic dimensions.
- Pronunciation
- Grammar
- Vocabulary
- Fluency (including response time and hesitation)
- Sentence complexity
Example insight
"Student performs at A2 level, but demonstrates B1-level sentence structure."
Benchmark vs. CEFR A2
Camille Moreau · Last 30 days
Insight
Performs at A2, but demonstrates B1-level sentence structure.
AI Learning Gap Detection.
The platform identifies recurring mistakes and learning gaps over time, surfacing patterns that would otherwise stay invisible between formal assessments.
- 17×
Frequent errors in verb conjugation (past tenses)
Grammar - 12×
Limited sentence complexity in written responses
Syntax - 9×
Repeated reliance on basic adjectives
Vocabulary
Five steps from everyday work to teacher insight.
- 01
Students continue their normal coursework
No new testing framework required.
- 02
The platform analyses existing recordings, assignments, and test responses
Recordings can be uploaded by the teacher or submitted directly by students through the Fluency EI mobile interface; an API connection is planned (coming soon).
- 03
Fluency EI evaluates each submission
Across pronunciation, grammar, vocabulary, fluency, and sentence complexity.
- 04
Teachers receive structured insights and level tracking
Available in student history and reports today. A progress report — a summary of student progress and areas needing attention, delivered as often as the school assesses — is planned, not yet sent automatically. Report scope and recipients will be confirmed before launch.
- 05
Students are grouped and supported more effectively
Cohort placement and intervention become data-driven.
How to use Fluency EI in your school.
Fluency EI is designed to work with your existing teaching methods. Whether you run a fully online program or a traditional classroom, the goal is the same: a clearer picture of every student's progress over time. Schools can choose from multiple simple implementation approaches depending on their setup.
Flexible implementation options
Suggested approaches, not fixed requirements. Each school chooses its own frequency — or combines several approaches.
Ongoing Analysis (Online Schools)
Uses existing recordings, assignments, and test responses. Fully passive — no additional work required with the planned API connection (coming soon). Recordings can be uploaded by the teacher or submitted directly by students through the Fluency EI mobile interface.
Weekly Speaking Checkpoint
Students submit a short 30–60 second spoken response. Can be done in class or at home.
Audio Homework Integration
Add one short speaking task to weekly homework. No classroom disruption.
Oral Exam & Presentation Analysis
Record existing oral exams or presentations. Fluency EI provides structured evaluation and benchmarking.
For online schools
- Uses existing student data
- Regular, automatic assessment
- CEFR level tracking over time
- Early visibility of learning gaps
“Everyday work, lasting insight.”
For classrooms and traditional schools
- Simple weekly speaking tasks
- Flexible integration into lessons or homework
- Clear progress data for every teacher
- No need for constant recording
“Minimal effort, powerful visibility.”
Works with Google Classroom
A low-tech, traditional school can use Fluency EI with the tools it already runs on — students never need a new account. Students can also submit directly through the optional Fluency EI mobile interface, which saves the teacher from collecting and uploading files. Guided onboarding and AI-assisted help are available for any school or teacher who would like extra support getting started, and during early deployments the Fluency EI team supports schools directly.
- 1Students record & submit
On a phone or Chromebook, students record a short spoken response and upload it to Google Classroom as they would any assignment.
- 2Teacher runs the analysis
The teacher downloads the recording from Classroom and uploads it directly to Fluency EI for pronunciation, fluency, grammar, vocabulary and sentence complexity. An API connection is planned (coming soon).
- 3Feedback stays in Classroom
The teacher copies scores, CEFR levels and remarks into a private comment in Classroom, or shares a printable report. Original submissions remain in Classroom; a copy is uploaded to Fluency EI for analysis.
The only new tool is the teacher’s own screen.
What this looks like in practice
A teacher asks students to record a short response to a question in the language they are learning. The platform analyses fluency, grammar, vocabulary, and pronunciation, and updates each student’s level automatically.
Built for modern schools.
Teacher decision support
Evidence that helps teachers decide — the teacher always has the final word.
Persistent student understanding
Replace point-in-time scores with a level signal built from each student's accumulating assessment history.
Better student placement and grouping
Cohorts informed by real data, not approximations.
Early detection of learning gaps
Surface patterns weeks before traditional reviews would.
Reduced teacher workload
Structured insights replace hours of manual evaluation.
Data-driven teaching approach
Move from intuition to evidence at the program level.
Built for education. Powered by evolving AI.
Fluency EI has been developed and fine-tuned over months with experienced language teachers to turn advances in artificial intelligence into a structured educational system.
The underlying AI technology is an engine — not the product. Fluency EI provides the educational methodology around it: structured language assessment, CEFR-aligned evaluation, teacher validation, persistent progress tracking and Educational Intelligence.
As AI models continue to evolve, Fluency EI can evolve with them. What remains constant is the educational framework, the assessment methodology, the student history and the teacher-led decision process.
Fluency EI proposes. The teacher decides.
The whole cohort, at a glance.
Each student's level, trajectory, strengths, and benchmark gap — updated as they learn.
FR-204 · Intermediate cohort
· 24 studentsUpdated 2 min ago
| Student | Level | Progress to next | Strength | Weakness | 30-day score change |
|---|---|---|---|---|---|
CM Camille Moreau | A2 → B1 | 62% | Vocabulary | Grammar · past tense | +8 |
LB Liam Bennett | B1 → B2 | 38% | Grammar | Pronunciation | +3 |
AD Aïcha Diallo | B2 → C1 | 71% | Fluency | Grammar · subjunctive | +12 |
NS Noah Schmidt | A1 → A2 | 84% | Sentence complexity | Vocabulary | +6 |
SR Sofia Romero | B1 → B2 | 22% | Pronunciation | Grammar · articles | −2 |
30-day score change: difference between each student's latest overall score and their score from 30 days earlier, on Fluency EI's internal 0–100 scale. Demo data shown for illustration.
Do you want to use Fluency EI in another language?
The interface language is the language the site is written in; the language assessed is the language the student speaks. They can differ — this English interface assesses French by default, and the flags above choose the language assessed. Schools working in another language can start from their own version.
Frequently asked questions
Could a general-purpose AI model assess a student's language?+
Yes. AI models are becoming increasingly capable of analysing spoken and written language. Fluency EI turns that capability into a consistent educational system designed specifically for teachers and schools — with structured assessment methodology, CEFR alignment, teacher oversight, student progress over time and Educational Intelligence.
The AI model is an engine. Fluency EI is the educational system around it.
What happens if a school or student wants to assess a different language or operates in a different country?+
Fluency EI currently assesses five languages: English, French, German, Spanish and Italian, each with its own speaking prompts and CEFR level expectations. Schools choose the language they teach, and the platform itself is available in English, with introductory pages in French, German, Spanish and Italian.
Because assessment is based on the CEFR — an international framework — schools in any country can use it for these five languages. Further languages can be added in future; we're happy to discuss your school's needs.
What happens to a student's assessment history if the student changes school?+
Today, a school can export a student's full assessment record on request; direct transfer between schools is on our roadmap.
How will each school's student data be protected?+
School-level data isolation is currently being implemented as part of Fluency EI's preparation for live school deployment.
Before any school is onboarded for live use with real student data, each school will have a secure environment in which access is restricted to its own authorised users and student data.
This control is currently work in progress and will be in place before live school onboarding.
We're a low-tech or Google Classroom school — will students need new software?+
No. Students only use their phones or Chromebooks to record audio and submit it to Google Classroom, exactly as they would any other assignment. They never need to create an account.
Optionally, students can instead record and submit directly through the Fluency EI mobile interface using a personal code from their teacher — no account, no app store. This saves the teacher from receiving, organising and uploading files, leaving more time for reviewing evidence and supporting students. Fluency EI proposes; the teacher decides.
The teacher downloads the recording from Classroom, uploads it directly to Fluency EI, and runs the analysis — pronunciation, fluency, grammar, vocabulary and sentence complexity. The teacher then copies the feedback into a private comment in Classroom, or shares a printable report.
Original submissions stay inside Classroom; a copy is uploaded to Fluency EI for analysis. Direct teacher upload and optional direct student submission are both available today; an API connection is planned (coming soon). Guided onboarding, AI-assisted help and, where required, direct support from the Fluency EI team are available.
Bring AI-powered intelligence to your language programs.
See how Fluency EI fits into your existing platform. We'll walk you through a sample cohort and tailor a pilot to your school.
30-minute walkthrough · No commitment