Guides · On camera
You cannot read a teleprompter without moving your eyes. Reading is made of small jumps: in ordinary reading, the eyes hop about seven to nine letters at a time and rest for a fifth to a quarter of a second between hops. What you can do is make the movement small, put it where viewers are least likely to notice it, and stop it looking like reading. Four things do that: distance, line width, where the text sits relative to the lens, and what the text does while you read it.
What a viewer sees is the angle your eyes sweep from the start of a line to its end. That angle is 2 × arctan(line width ÷ (2 × distance)). A line 30 cm wide, read from 60 cm away, spans about 28 degrees. The same line from 1.2 m spans about 14. That is why the usual advice is to move back from the prompter, and for a camera on a tripod it is good advice.
At a desk you cannot move back far, so narrow the line instead. A 12 cm column read from 60 cm spans about 11 degrees. Make the column narrow first, then set the type only as large as you need to read it comfortably.
Viewers are not equally sensitive in every direction. A 2002 study of videoconferencing at Stanford found people an order of magnitude less sensitive to a gaze directed below their eyes than to one directed to the left, to the right or above. The study was about video calls, not prompters, and applying it here is our inference: text a little under the lens is the offset viewers are least likely to notice, and text beside the lens is more likely to read as looking away.
With a laptop webcam, that means the top centre of the screen, in short lines. At 60 cm, text 5 cm below the lens is about 5 degrees off it.
Pick a reading line about a third of the way down the text area and let the text come up into it, instead of chasing lines down the screen. Your eyes then make small sideways movements at one height, without searching up and down.
Use short sentences. Break lines at the end of a phrase, so each line is one thought you can take in at a glance and then say to the viewer. Read the script aloud twice before the real thing. By the third read you need the page less, and your eyes show it.
Eyes that sweep left to right and snap back are reading, and in our view viewers recognise it without being able to say why. The opposite failure is the fixed stare. Pin the script under the lens and your eyes are steady but locked on one point, and nobody holds one point for a whole meeting except someone reading from it. The goal is not zero movement. It is movement that looks like a person talking to people.
Terera places the script for the audience you choose. For one person, it keeps the line near your camera, with small natural drift. For two people, it holds the text in two fixed panels, which you place by the faces on your call; for a room, Stations holds it at fixed positions. In both, the text hands over at the end of a sentence, so nothing slides while you are reading it. Two people means two listeners, not two presenters.
A scrolling teleprompter moves at a set speed, and you keep up with it. Terera follows your voice word by word, waits when you pause, holds your place when you go off script, and resumes when you return to the script.
It does not stop your eyes moving, and it does not track them. It puts the words where your eyes would go anyway. The home page has a simulated demo that does not use your microphone.
Terera
A teleprompter for Windows and Mac. USD 89 once, tax included. Three of your own machines, 12 months of updates, and the installed version keeps working after that. A 30-day refund by email. Speech recognition runs on your own computer.
No. Reading happens in small jumps, about seven to nine letters at a time in ordinary reading. You can make the movement small by narrowing the lines or sitting further back, and keep it just below the lens, where viewers notice it least.
Far enough that a line of text spans a small angle. A 30 cm line spans about 28 degrees from 60 cm and about 14 degrees from 1.2 m. At a desk, narrow the column instead: a 12 cm column read from 60 cm spans about 11 degrees.
Just below the camera, in short lines. A 2002 videoconferencing study found viewers an order of magnitude less sensitive to a gaze directed below their eyes than to one directed to the side or above.
In our view, often. The usual tells are a left-to-right sweep across wide lines and a fixed stare at one point.
No. Terera follows your voice word by word and places the script where your eyes would go for the audience you chose: one person, two people or a room. It does not track your eyes, and it never sees your call.
Related guides: a teleprompter for Zoom, a teleprompter for Microsoft Teams and a teleprompter that follows your voice word by word. What leaves your computer: Security. Compared with other teleprompters: the dated comparisons. All guides.