In film, every sound in a video falls into one of two buckets.
Diegetic sound lives inside the world of the story. The characters could hear it if they were paying attention. A car driving past on screen, a phone ringing in the same room, a song playing on a radio in the scene. The source of the sound is part of the story.
Non-diegetic sound lives outside the world of the story. The characters can’t hear it, because it isn’t really in their world. The score that swells during an emotional moment, the narrator’s voice describing the action, a sound effect added later for emphasis. The audience hears it. The characters don’t.
The word “diegetic” comes from the Greek “diegesis,” meaning “narrative” or “the world of the story.” Diegetic sound is part of that world. Non-diegetic sound is layered on top by the filmmaker.
Here are a few examples:
This split matters for understanding how sound design works.
Non-diegetic sound is one of the most important tools for shaping how an audience feels.
Music for emotion. A scene with no music feels different from the same scene with a swelling score. Music tells the audience how to feel: tense, romantic, triumphant, sad. Non-diegetic music is the most common way films steer emotion.
Narration for context. A voice-over narrator can explain the plot, share what a character is thinking, comment on the action, or move the story along without showing it. Documentaries lean on this heavily. Some narrative films use it as a stylistic choice.
Sound effects for emphasis. A subtle whoosh under a transition, a sharp sting at a big reveal, a bass drop on an impact. None of these exist in the characters’ world, but they shape how the moment lands for the audience.
Bridging cuts. Non-diegetic music that carries across cuts ties together scenes that would otherwise feel disconnected. A song under a montage links the separate shots into one sequence.
Setting the tone. A low non-diegetic drone under a scene can suggest dread, mystery, or unease even when nothing specific is happening on screen.
Signalling genre. Certain sound choices tell you the genre before anything else does. Low, sinister frequencies suggest horror. Bright orchestral chords suggest adventure. Synth-driven music suggests sci-fi or thriller. The audience reads these signals before any plot details.
For commercial and online video, non-diegetic sound is essential. The music track on a brand film, the sound effects on a motion graphic, the voice-over on a tutorial: all non-diegetic. Without them, most modern video would feel flat.
Some films use non-diegetic sound in ways that define their whole feel.
Star Wars. John Williams’ score is heavily non-diegetic. The audience isn’t hearing what the characters hear. They’re hearing an orchestral commentary on the action. The Imperial March, the Force theme, and Princess Leia’s theme are all non-diegetic motifs that shape how you read the story.
The Godfather. Nino Rota’s score is non-diegetic, mournful, and full of feeling. You feel the weight of the Corleone family’s choices partly because the music tells you how to feel.
Apocalypse Now. The famous helicopter assault uses Wagner’s “Ride of the Valkyries” as both diegetic (the helicopters play it through speakers) and non-diegetic (it scores the scene emotionally). That double role is part of why the scene works.
Documentary narration. From David Attenborough’s BBC nature films to Errol Morris’s documentaries, non-diegetic narration shapes how the audience reads the images.
Wes Anderson films. These often use unexpected non-diegetic music, like vintage pop songs or baroque chamber music, as a key part of the style. The music feels deliberately like a soundtrack choice, not a coincidence, and that’s the point.
For commercial work, non-diegetic music is the default. A brand film with no soundtrack feels stark and bare. The same film with a strong music track feels polished and emotional. Choosing the right non-diegetic sound is often the single biggest creative decision in commercial post-production.
Some of the most interesting sound design plays with the line between diegetic and non-diegetic.
Music that “becomes” diegetic. A scene starts with non-diegetic music, where we hear the soundtrack but see no music source. Then the camera pans to reveal a radio, a record player, or a band, and it turns out the music was coming from the scene all along. Films use this trick for a fun surprise.
Music that “becomes” non-diegetic. The opposite. A character starts humming or playing music in the scene (diegetic). Then the music swells and becomes the soundtrack, carrying across cuts and into scenes where the original source isn’t there.
Internal monologue. A character’s thoughts heard as voice-over. Technically non-diegetic, since the other characters can’t hear it, but coming from a character inside the scene. The line is fuzzy.
Sound effects only the audience hears. A subtle whoosh on a quick cut, a bass drop on an impact, a sting on a reveal. None of these exist in the story world, but they’re central to how the moment feels.
Fourth-wall-breaking moments. When a character speaks to the camera and the music suddenly stops, that change in non-diegetic sound is itself a storytelling choice. It tells the audience the story has shifted into a different mode.
Good sound designers think carefully about these boundaries. A film with sloppy diegetic and non-diegetic choices feels muddled. A film that plays with the boundary on purpose can create some of the most memorable moments in cinema.
In commercial and online video, almost every sound layer is non-diegetic: the music, the effects on motion graphics, the voice-over. A dedicated editor treats that layer as creative space, not just a polish step, because the right music shapes how viewers feel and the right effects tell them what to notice. Choosing those sounds with care, rather than dropping in the first track that fits, is a normal part of how we finish a video.
Diegetic sound lives inside the world of the story, where the characters can hear it: a phone ringing in the room, a radio playing in the scene. Non-diegetic sound lives outside the story, where only the audience hears it: soundtrack music, narration, sound effects added in post-production. The split matters for understanding how sound design shapes what the audience feels.
Usually yes. The orchestra isn't physically in the scene with the characters, so the soundtrack is layered on by the filmmaker and is non-diegetic. The exception is when a song playing in the scene, like on a radio, at a concert, or from a record player, also works as the soundtrack. That song is diegetic and acts as soundtrack at the same time, so the line blurs.
Almost always. A narrator describing the action from outside the story is non-diegetic. A character's voice-over from outside the action they're describing, like a future reflection or an internal monologue, is also non-diegetic. The rare exception is a character whose narration is part of the story itself, where they're speaking directly to other characters in the scene. That's diegetic.
It depends on the sound. Footsteps, doors closing, and car engines are diegetic, because they stand for sounds in the story world, even if foley artists made them in a studio. Stings, whooshes on cuts, and low drones under a scene are non-diegetic, because they don't stand for any source in the story. Most films use both layers.
It helps editors and sound designers think clearly about what each sound is doing. Diegetic sounds make the world feel real. Non-diegetic sounds shape how the audience feels. Knowing which layer a sound belongs to helps you mix it the right way. It also opens up creative possibilities for crossing the line on purpose, since moments where diegetic music becomes non-diegetic, or the other way around, can be powerful storytelling tools.
