Skip to content

Classroom observation

Multimodal Utterance Recognition × Reflection

Classroom conversations, made measurable.

MUR² turns a recorded class discussion into plain measurements of participation: who spoke, for how long, and how long the room waited before someone answered. Then it helps you reflect on them.

MUR² observes. MUR² measures. MUR² helps educators reflect. It does not grade teachers.

01 / Measures

Deterministic · Explainable

What it measures

Participation, in numbers you can explain. They come from ordinary, unit-tested code. A language model never does the arithmetic; it only helps put the results into words.

M1

Talk share

How much of the active discussion time the instructor and the students each spoke, as a percentage.

M2

Unique speakers

How many different anonymous voices took a real turn: at least 2.0 seconds of speech, and you can change that threshold.

M3

Concentration

Whether a few voices carry the room: a Gini score, plus the share held by the top four speakers.

M4

Student to student

Turns that pass from one student straight to another. This is the strongest sign of real discussion.

M5

Wait time

How long the room waited after a question before someone responded. Groups with fewer than five samples are left out, not guessed.

M6

Any stretch of class

Brush a stretch of the timeline and every measure is recomputed for just that window.

Example reportIllustration, not real data
  • Instructor spoke for 54% of active discussion time.
  • 13 anonymous speakers participated.
  • 4 students accounted for 61% of student speech.
  • Median wait time was 2.1 seconds.
  • Remote students contributed 18% of student turns.

Every number can be explained

Every number in a report is built to be explainable. When the audio is too unclear to trust a measure, MUR² leaves that measure out instead of guessing.

02 / How it works

Capture · Recognize · Understand · Measure · Reflect · Improve

From recording to reflection

Follow a recording from classroom audio to a report you can review.

  1. 01

    Capture

    Start with a recording of a class discussion.

  2. 02

    Recognize

    Speech models estimate who spoke when, and what was said.

  3. 03

    Understand

    MUR² turns that raw speech into turns, questions and pauses.

  4. 04

    Measure

    Deterministic, unit-tested code computes every number.

  5. 05

    Reflect

    You predict before class, then compare your prediction with what was measured.

  6. 06

    Improve

    Take one question into the next session.

03 / Slides

Speech · Screen · Timeline

Slides and speech, side by side

When a screen is shared, MUR² reads the slides on your own device with Apple’s Vision framework, then deletes the screenshots. Only the slide titles, their text and the times are kept.

Those slide words go to the speech recognizer as a glossary, so subject terms come out spelled the way your slides spell them.

The report then shows the slides beside the Murmur Timeline: one lane per speaker, layers you can switch on and off, and a brush to zoom into any stretch. Low-confidence stretches are hatched rather than tinted, and colour is never the only cue.

A slide beside its timelineIllustration, not real data

Slide 1

Introduction

Slide 2

Heteroscedasticity

Slide 3

Worked example

Glossary from slide text

…the residuals show heteroscedasticity, so the usual assumption fails…

S1S2S3S4SLIDE 1SLIDE 2SLIDE 3

04 / Reflection

Observe · Measure · Reflect

Reflection, not evaluation

MUR² observes. MUR² measures. MUR² helps educators reflect. It does not grade teachers.

There is always a report: with no API keys at all, a built-in fallback writes the plain-language summary. When a language model is used, it only narrates numbers that code has already computed.

  1. 01

    Predict

    Before class, commit to a prediction of how the discussion will go.

  2. 02

    Estimate

    Before the results appear, estimate what the numbers will show.

  3. 03

    Compare

    See your prediction, your estimate and the measured result side by side.

  4. 04

    One question

    Leave with a single question to carry into the next session.

05 / Privacy

Private by default

Private by default

A classroom is not a data source to be mined. These limits are part of the design, not settings to hunt for.

What it does

  • Audio is deleted once it has been processed.
  • Speakers are anonymous, and are never linked from one session to the next.
  • Students do not need accounts.
  • Transcripts are redacted by default: emails, phone numbers, links and IDs are removed.
  • Turn transcript retention off and only counts are kept.
  • Deleting a session deletes everything about it.
  • Session data lives in a local database on the laptop that ran the engine.

What it will never do

  • Camera-based emotion recognition
  • Identifying or grading individual students
  • Evaluating or ranking faculty
  • Covert recording: MUR² is for classes that know they are being recorded
  • Proctoring or gaze tracking
  • Selling or sharing classroom data

Read the privacy policy

06 / Status

v1.4.0 · Tested, not yet validated

Where it stands

Version 1.4.0. Built and tested, and honest about what is still unproven.

In place and tested

  • The metrics engine
  • The recording-to-report pipeline
  • The engine’s API
  • Class mode and website mode
  • Slide reading
  • The web interface

Not yet validated

What MUR² reports has not yet been checked against a hand-labelled class. Treat every number as a measurement with unknown error until it has.

The checks for speakers, talk time, questions and wait time are built and ready. Telling voices apart from across a real classroom is hard, and no public benchmark exists for it.

Inspired by Loquat (Lachko Hristov, RTI International, CIES 2024). MUR² is an independent implementation and is not affiliated with RTI.

07 / Contact

Try it on one class

Open the dashboard to add a recording and read its report.