What a self tape needs to get right.

There are two layers to a self tape. The first is the casting request: the sides, the slate instructions, the framing notes, the deadline, the naming pattern. Those are the actual requirements, and they override everything else, including everything on this page. The second layer is convention, the set of defaults the industry has settled on for what a tape looks and sounds like when nobody specifies otherwise. This page is the second layer. Read the request all the way through before you set up a single light, then use these defaults to fill whatever gaps it leaves open.

Framing and eyeline

The default frame is a medium close-up. Bottom of frame around the chest, top of frame just above the head with a little air, so you have room to breathe without clipping. Your eyes sit in the upper third of the picture. Landscape orientation, always, unless the request says vertical.

Put the camera at eye height. A camera looking up at you distorts the face and reads as looming; a camera looking down shrinks you. Eye height is neutral, and neutral is what you want from the equipment.

Your eyeline goes just off the lens, to whichever side your reader is standing, close enough to the camera that casting can see both eyes working. Do not look straight down the barrel during the scene; that is for the slate, if there is one. The frame matters more than it feels like it should, because casting forms an impression before you finish your first line. The first ten seconds breaks down exactly what they are watching for.

The picture: background and light

Plain, uncluttered, neutral. A solid wall, a hung backdrop, a stretch of clean space in a flat color that is not fighting your face. No bookshelves, no open doors, no laundry, and no windows behind you, because a bright window turns you into a silhouette and pulls the exposure away from your eyes.

Light the face evenly. The simplest working setup is a soft light source in front of you, slightly above eye level, with you standing far enough off the wall that you do not throw a hard shadow onto it. Avoid lighting from overhead fixtures alone; they hollow out the eyes. You should be a touch brighter than the background, so the eye goes to you without thinking about it.

Wardrobe follows the same logic: a solid color that separates you from the wall, nothing busy, nothing with a logo.

Sound

Casting will forgive an imperfect picture long before they forgive audio they have to strain against. Kill the fan, the air conditioner, the open window over traffic. Record at a conversational distance from the phone or mic, in a room that does not echo; soft furniture and closed doors do most of that work.

The balance to hit: you are clearly the primary voice, and the reader is clearly audible underneath you. A reader casting cannot hear makes the scene impossible to follow. A reader louder than you makes it their tape. Position them near the lens on your eyeline side, close enough to be picked up, quieter than you but never mumbling. Choosing a reader covers what to listen for in the voice across from you, and if nobody is available at all, the self-tape without a reader guide covers that setup honestly.

The slate

The slate is the one element you never freelance. If the request asks for one, deliver exactly what it lists, in the order it lists, at the position it names, head or tail. If the request says nothing about a slate, do not invent one. A slate nobody asked for, or a slate missing the detail they wanted, announces that you skimmed the instructions before they have seen a second of the scene. Slating without apologizing covers how to deliver one when it is called for.

Files and naming

One scene per file, unless the request says to combine them. Each scene is a single continuous take; do not cut inside the scene. Export a standard format like MP4, at a sensible size for the length, and check the request or the platform's help pages for any stated limits rather than trusting a number you heard once. Then play the export back before you upload. All the way through, with the sound on.

Name the file the way the request specifies, character for character. If no pattern is given, a clean name with your full name and the role is the safe default. The upload mechanics differ by platform; the Actors Access, Casting Networks, and Eco Cast guides each walk the specific flow, and self-tape turnaround covers doing all of this on a short clock.

When the request is silent

Here is the hierarchy, one more time, because it is the whole page. The request first. These defaults second. Your habits last. When casting writes an instruction, that instruction is the requirement, however odd it seems, and following it precisely is part of the audition. When the request says nothing, the defaults above are what silence means: medium close-up, landscape, eye height, plain background, even light, clean audio with the reader under you, no slate unless asked, one scene per file, a standard export, a clear name.

Get the specs set once and they stop costing you thought. That is the point of them. The variable left in the tape is the performance, and that is decided before you press record, by the rehearsal behind it. The full prep sequence walks from cold sides to a submission-ready take.