School

The School tools are made for the classroom itself. They run in the browser on whatever drives the projector or the interactive whiteboard, so there is nothing to install. The first is the classroom noise meter: a light that turns green, amber or red as the room gets louder, against a quiet limit chosen for the activity.

• Starting limits run from 50 dB(A) for silent work or a test to 80 dB(A) for practical work or play, and any other limit can be typed in.
• Each session keeps its average level, its highest level, how often the light turned red and the longest quiet stretch.
• Calibrated against a real sound level meter, the estimated decibels become readings for that device.
• A daily noise exposure panel shows where a teacher's working day stands against the values of the Control of Noise at Work Regulations 2005.

The story dice roll up to 9 picture dice from 6 picture sets, and a pupil, a group or the whole class tells or writes a story that uses every picture. The national curriculum in England asks pupils to give narratives in spoken language and to plan their writing by saying out loud what they will write, and the dice give that talk a starting point.

Players can take turns on one story, beginning, middle and end, with a timer per turn, a story starter and the picture names shown under the dice. The same roll can be sent to a class as a link, and each die prints as a cube net with a 3.5 cm edge to make at home.

Much of a lesson is learned by listening, and speech is understood only when it stands clearly above the sound around it. The World Health Organization's 1999 guidelines on community noise ask for speech to be at least 15 dB louder than the background for complicated messages at school.

Shield, Greenland and Dockrell's 2010 review of research on noise in schools adds that the youngest classes need a wider margin than older pupils. Lessons are rarely that quiet: in London primary schools, Shield and Dockrell (2004) measured occupied classrooms at 72.1 dB LAeq on average, against 47 dB when the same rooms stood empty.

Noise changes the speaker as well as the listener. In a louder room voices rise without anyone deciding to, which is known as the Lombard effect. Bottalico and colleagues (2017) measured a rise of about 0.65 dB for each decibel of noise once the room passes about 43.3 dB(A), so a class that stays loud keeps the teacher's voice raised for longer.

A quiet limit only works when it matches what the lesson asks of the class. Silent work needs a far lower limit than a science practical, so the noise meter starts from the activity rather than from one figure for the whole day. Five starting limits cover the usual kinds of lesson, and each can be changed.

Shield, Greenland and Dockrell's 2010 review of research on noise in schools reports typical levels for each kind of activity in enclosed primary classrooms. Each starting limit sits at or a few decibels above that level, so a class working as planned keeps the light green, and red marks a room that has gone past the plan.

• Silent work or a test: limit 50 dB(A), where the review reports 44 dB LAeq with pupils silent.
• Quiet reading: limit 60 dB(A), where quiet activities such as silent reading come to 56 dB LAeq.
• Individual work with some talking, the default: limit 68 dB(A), against 65 dB LAeq in the review.
• Group work: limit 76 dB(A), against 70 to 77 dB LAeq in the review.
• Practical work or play: limit 80 dB(A), the site's own choice, since the review gives no figure for it.

The limit can also come from the room itself. Pressed while the class works at a level the teacher accepts, the room button sets the limit 3 dB above the steady level measured at that moment. The limit then rests on the same microphone that measures, which helps where the device has not been calibrated.

Two settings soften the light. The amber band begins 5 dB below the limit by default, as a warning before red. The steady response follows the median level of the last 5 s, so a cough, a dropped book or a scraped chair does not turn the light red; only a room that stays loud does.

The noise meter works on a laptop, a tablet, a phone or the computer behind a projector or an interactive whiteboard. In full screen it shows only the light and its word on a dark stage, and the screen is kept awake while it measures where the browser allows that.

It needs the microphone, so the browser asks for permission the first time, and only on a secure page. The sound is turned into levels inside the page: nothing is recorded, uploaded or kept. A calibration, once saved, stays in that browser, and a shared link carries only the settings.

A built-in microphone hears the room from where the device stands, so a laptop on the teacher's desk hears the nearest pupils loudest.