Run #775

done rubric v1 session 79416979 gemini-3.7-flash 2026-09-21 02:21

Transcript

Not used by the HEARD criteria — Only 157 characters of speech after cleaning (200 were the language-detection artefact) — too little to judge teaching from.

Read it no speaker labels — tutor and student run together

Hello Sachin, welcome to Filo. Good evening. Good evening. Sachin, आपनी question पूछा है, what color does acid change on pH paper? क्या यही question है आपका?

Transcript

Not used by the HEARD criteria — Only 157 characters of speech after cleaning (200 were the language-detection artefact) — too little to judge teaching from.

Read it no speaker labels — tutor and student run together

Review team

no_issue

Product AI

nothing found

teaching Yes · mistakes No · satisfactory Yes

Each verdict takes your own. Yours is stored against the stack's, so “where is it wrong most often” is a question with an answer.

Work is the right way up The video feed is upside down.

Across frames 2–7, all text (e.g. 'Acid -> PH value' and Hindi handwriting) is inverted at 180 degrees orientation.

seen · 100% confident
Your view
Writing is legible at video size Handwriting is clear and legible despite orientation.

In frames 6 and 7, text such as 'Acid -> PH value (0-7)', 'strong acid, weak acid', and 'Blue litmus paper -> Red litmus paper' is clearly readable.

seen · 90% confident
Your view
The work fills the frame Notebook page fills most of the frame.

In frames 2–7, the spiral notepad is centered and occupies the vast majority of the camera view.

seen · 95% confident
Your view
Nothing distracting in shot Workspace is clean with minimal surroundings visible.

Frames 2–7 show only the notebook, the tutor's hand, a pen, and the dark desk surface on the borders.

seen · 95% confident
Your view
The student's problem is established No clear problem statement or original question is displayed.

Frames 2–7 show notes being taken about acids and pH, but the original question or prompt being answered is not visible.

seen · 80% confident
Your view
Work proceeds in followable steps Work progresses logically across frames.

Frames 3 through 7 show sequential additions: first the pH range, then the pH scale line, and finally the litmus test effect.

seen · 90% confident
Your view
It reaches an answer The explanatory notes are completed.

By frame 7, the explanation covering acid pH range, scale markings, and litmus paper reaction is fully written out.

seen · 85% confident
Your view
Something was actually written Not applicable to a camera session

measured
The concept is named before the working Not assessed

Only 157 characters of speech after cleaning (200 were the language-detection artefact) — too little to judge teaching from.

heard
Your view
Explained rather than jumped to the answer Not assessed

Only 157 characters of speech after cleaning (200 were the language-detection artefact) — too little to judge teaching from.

heard
Your view
Solved step by step, not in one leap Not assessed

Only 157 characters of speech after cleaning (200 were the language-detection artefact) — too little to judge teaching from.

heard
Your view
The session opens properly Not assessed

Only 157 characters of speech after cleaning (200 were the language-detection artefact) — too little to judge teaching from.

heard
Your view
No long dead air 0 gaps over 6.0s, 0s total (0% of the session), longest 0s

Silence is below -35.0dB — a crude measure on a phone microphone, so treat a low count as weak evidence rather than proof the tutor was talking.

measured
Your view

Observations judged, but not counted as the lab objecting

The camera holds still picture moves 37.6 on average between keyframes, 0.6 hard jumps/min

Consistent with the camera being handheld — a stand would remove this entirely. A handheld phone measured 35.1. No steady-camera session has been measured yet, so the lower band is a guess.

Not counted: Measured over 183 known-bad and 29 known-good camera sessions (2026-09-21): raised on 87% of the bad and 86% of the good, and the motion distributions are the same shape — median frame difference 25 on the bad set, 27 on the good. No threshold on this number separates them, so it is reported and not counted.

measured
Your view
Exposure is usable average brightness 118.8, 34.0% of frame very dark, 13.0% blown out

Below 70 average, or over 25.0% of the frame too dark to read.

Not counted: Same two sets: it raised 6 of 183 known-bad and 2 of 29 known-good sessions. Both too few to mean anything, and the brightness distributions overlap almost exactly. The earlier reading that it 'fires more on clean sessions' was two sessions out of twenty-nine — noise, not an inverted rule.

measured
Your view
Understanding of the concept is checked Not assessed

Only 157 characters of speech after cleaning (200 were the language-detection artefact) — too little to judge teaching from.

Not counted: Raised on 87% of known-bad and 80% of known-good sessions. The criterion is honest about why in its own `why`: the transcript has no speaker labels, so a real check and a rhetorical 'ok?' are often indistinguishable, and the bar ends up one almost no session clears. Still judged and shown, because reading it beside a recording is how it gets better.

heard
Your view

What the model returned verbatim

{
  "findings": [
    {"key": "orientation", "verdict": "bad", "headline": "The video feed is upside down.", "evidence": "Across frames 2–7, all text (e.g. 'Acid -> PH value' and Hindi handwriting) is inverted at 180 degrees orientation.", "confidence": 1.0},
    {"key": "legibility", "verdict": "ok", "headline": "Handwriting is clear and legible despite orientation.", "evidence": "In frames 6 and 7, text such as 'Acid -> PH value (0-7)', 'strong acid, weak acid', and 'Blue litmus paper -> Red litmus paper' is clearly readable.", "confidence": 0.9},
    {"key": "framing", "verdict": "ok", "headline": "Notebook page fills most of the frame.", "evidence": "In frames 2–7, the spiral notepad is centered and occupies the vast majority of the camera view.", "confidence": 0.95},
    {"key": "workspace", "verdict": "ok", "headline": "Workspace is clean with minimal surroundings visible.", "evidence": "Frames 2–7 show only the notebook, the tutor's hand, a pen, and the dark desk surface on the borders.", "confidence": 0.95},
    {"key": "question_captured", "verdict": "weak", "headline": "No clear problem statement or original question is displayed.", "evidence": "Frames 2–7 show notes being taken about acids and pH, but the original question or prompt being answered is not visible.", "confidence": 0.8},
    {"key": "structure", "verdict": "ok", "headline": "Work progresses logically across frames.", "evidence": "Frames 3 through 7 show sequential additions: first the pH range, then the pH scale line, and finally the litmus test effect.", "confidence": 0.9},
    {"key": "completion", "verdict": "ok", "headline": "The explanatory notes are completed.", "evidence": "By frame 7, the explanation covering acid pH range, scale markings, and litmus paper reaction is fully written out.", "confidence": 0.85}
  ]
}

Duration	96s (1.6 min)
Video	960×540 @ 15.0fps
Audio track	yes
Shot changes
Dead air	0 gaps ≥6.0s · 0s total · longest 0.0s
Luma	avg 118.8 · min · max
Medium	camera 29.7% flat background
Content written	9.0% → 53.77% (+61.58)
Transcript	157 characters of speech
Frames	7

Run #775

The recording

What the student sent the question as it arrived, then each reading of it

Transcript

What the student sent the question as it arrived, then each reading of it

Transcript

Review team

Product AI

Measured

What the model returned verbatim

Frames in order, one every ~14s