SPEAKCLEAR 7.1 — SPEAKCLEAR 7.1 Independent Research Build

System Workspace

Functional Build · Version 7.1

SPEAKCLEAR™

AI-Assisted Oral Assessment & CEFR Growth System
CEFR standards-aligned oral analysis, full criterion moderation, and rubric-aligned examiner rationale.

Criterion ACriterion B1Criterion B2Criterion CHuman Moderation

Assessment principle

Final decisions should be grounded in the relevant criterion. The app provides provisional evidence; the moderator confirms or adjusts using examiner rationale aligned to the construct of each criterion.

The SPEAKCLEAR Framework™

Ten-Point Oral Preparation System

SPEAKCLEAR helps students remember WHAT to talk about.

SPEAKCLEAR Memory Sentence

Smart Penguins Eat Angry Kangaroos, Creating Loud Emotional Alien Reflections.

Letter	Memory Word	SPEAKCLEAR Meaning
S	Smart	Say the main idea
P	Penguins	People
E	Eat	Evidence
A	Angry	Analysis
K	Kangaroos	Key words and phrases
C	Creating	Connections
L	Loud	Layout
E	Emotional	Emotion
A	Alien	Author's message
R	Reflections	Reflection

SPEAKCLEAR Model Sentences

“The author uses dark visual imagery and crowded panel layout to create fear and communicate the message that discrimination destroys human identity.”

Analysis Layout Emotion Author's message

“The key phrase ‘trapped like animals’ connects to the visual symbolism of the mice and reflects the author’s message about dehumanization during the Holocaust.”

Key words and phrases Connections Analysis Author's message

Transcript Workspace

Candidate 01, DP English B HL

Optional Live Speech-to-Text

Speak Live into the Transcript Box

This is live browser dictation, not uploaded audio-file transcription. It does not store audio in SPEAKCLEAR.

Live speech is off.

Privacy note: live speech recognition depends on your browser and microphone permission. SPEAKCLEAR does not upload or store audio files. Use anonymous candidate labels only. For official assessment use, follow school privacy policy.

Mode

Presentation or Full IOA?

ModeCandidate

B2 and C require follow-up discussion evidence. If the transcript is presentation-only, B2 and C should not be securely awarded.

IB Calibration Dashboard

Provisional Evidence + Moderated Final Marks

A Language—

Provisional

B1 Presentation—

Provisional

B2 Conversation—

Optional

C Interaction—

Optional

Final Total—

Awaiting moderation

Working Visualizations

Criterion Radar + SPEAKCLEAR Radar + CEFR Bar

SPEAKCLEAR Mnemonic

A · B1 · B2 · C

CEFR Growth

IB Calibration & Full Moderation

Human Moderation for Criteria A, B1, B2 and C

Each final mark must be supported by criterion-specific evidence. Select examiner rationale and write an examiner-style rationale.

Criterion A — Language /12

Awaiting analysis

Range, accuracy, pronunciation, intonation, fluency and intelligibility across the oral.

Final A

Criterion B1 — Message: Presentation /6

Awaiting analysis

Relevance, development and support of ideas in relation to the selected extract.

Final B1

Criterion B2 — Message: Conversation /6

Awaiting evidence

Relevance, development and appropriateness of answers in the follow-up/general discussion.

Final B2

Criterion C — Interactive Skills /6

Awaiting evidence

Comprehension, response, initiative, repair, turn-taking and ability to sustain the exchange.

Final C

Final A
—

Final B1
—

Final B2
—

Final C
—

Total
—

Run analysis first. Then moderate any criterion where human evidence changes the best-fit judgement.

Report Export Center

Download Local Student Reports

Reports are generated locally in your browser. Nothing is stored online by this app. Use anonymous candidate labels only.

Student HTML Report

Downloads a clean, anonymous report with final marks, moderation notes, examiner rationale, and growth feedback.

Print / Save as PDF

Opens a print-ready report window. Use your browser’s print menu to save as PDF.

Anonymized CSV Data

Exports only anonymous marks and moderation metadata for later local analytics.

Run an assessment and apply moderation before downloading final reports.

Student-Facing Rubrics

Full Visible Rubric Logic

Criterion A — Language /12

Band	Descriptor	Fit

Criterion B1 — Presentation /6

Band	Descriptor	Fit

Criterion B2 — Conversation /6

Band	Descriptor	Fit

Criterion C — Interactive Skills /6

Band	Descriptor	Fit

Growth Feedback

Performance Growth & Upgrade Analysis

Run analysis to generate feedback.

Communicative Authenticity & Delivery Moderation

Human-in-the-Loop Oral Delivery Evaluation

Protocol Reminder: Students should rely on brief 10-point preparation notes rather than fully scripted responses. The goal of oral assessment is spontaneous, natural communication. Flattened intonation, heavy script dependence, and over-rehearsed delivery may reduce communicative authenticity and interaction quality.

Delivery & Authenticity Indicators
Natural pacing Intonation variation Script dependence Communicative spontaneity Lexical authenticity

SPEAKCLEAR Delivery Flags™

Script Dependence Risk

Intonation Flattening

Lexical-Proficiency Mismatch

Interactional Fragility Risk

Communicative Authenticity Concern

The evaluator may document how vocal delivery influences Criterion A and later interactional performance. This panel evaluates communicative consequences rather than punishing students for scripting.

Examiner Insight

A candidate may demonstrate sophisticated vocabulary and strong extract analysis while simultaneously limiting communicative authenticity through monotone or heavily scripted delivery. Lexical sophistication alone does not guarantee upper-band Criterion A performance if intonation fails to enhance communication.

SPEAKCLEAR Framework & Philosophy

SPEAKCLEAR™ Framework & Philosophy

SPEAKCLEAR™

AI-Assisted Oral Assessment & CEFR Growth System

Version 7.1 · UNIFIED BRANDING & IDENTITY BUILD

Official Product Framework

SPEAKCLEAR™ is a premium academic assessment framework designed for oral communication, CEFR growth mapping, examiner calibration, multimodal extract analysis, and human-in-the-loop moderation.

Research Framework

Developed by Thomas J. Baker as a doctoral research prototype for pilot implementation, moderation analysis, communicative authenticity evaluation, and oral assessment innovation.

Visual & Branding Philosophy

SPEAKCLEAR maintains a unified visual framework across all builds through consistent typography, dark-blue academic visual architecture, CEFR growth symbolism, moderation-centered analytics, and coherent assessment language.

Assessment Philosophy

SPEAKCLEAR evaluates communicative performance through a balanced integration of language, message, interaction, multimodal analysis, delivery authenticity, examiner calibration, and growth-oriented feedback.

Independent Educational Research Framework

Legal & Institutional Disclaimer

SPEAKCLEAR™ is an independent educational research and assessment framework developed by Thomas J. Baker.

The system is designed for communicative oral analysis, moderation support, CEFR growth mapping, and educational research applications.

SPEAKCLEAR™ is not affiliated with, endorsed by, certified by, or officially connected to the International Baccalaureate Organization (IBO).

All assessment language, moderation commentary, and rubric-aligned examiner rationale contained within SPEAKCLEAR™ are independently developed for educational and research purposes.

References to international assessment standards, communicative competence frameworks, or oral assessment methodologies are used exclusively for educational alignment and research-informed calibration purposes.