title: Voice Actor slug: voice-actor aliases: - Voice-Over Artist - Voice Talent - VO Artist - Dubbing Artist category: Entertainment tags: - voice-over - acting - audio - performance - microphone difficulty: advanced summary: >- Acts with the voice alone, reading intention off the page and putting all character and physicality into sound, clean enough to edit and repeatable across sessions. contributors: - soul-atlas last_reviewed: null provenance: ai-generated created: '2026-06-26' updated: '2026-06-26' related: - slug: actor type: related note: shares the core craft of intention and direction, minus face and body - slug: sound-engineer type: collaboration note: turns the performance into a clean, usable signal - slug: animator type: adjacent note: supplies the body and timing the voice must match - slug: film-editor type: collaboration note: cuts the takes; depends on clean breaths and consistent levels - slug: musician type: adjacent note: shares the discipline of an instrument needing warmup and lifelong care - slug: broadcast-journalist type: adjacent note: works the same microphone and conversational baseline specializations: - Animation Voice Actor - Audiobook Narrator - Commercial Voice-Over Artist - ADR / Dubbing Artist country_variants: [] sources: - title: Word of Mouth kind: book - title: There's Money Where Your Mouth Is kind: book - title: The Art of Voice Acting kind: book status: draft reviewers: [] sections: - heading: Purpose markdown: >- A voice actor exists to make a listener believe in a person they will never see. Everything an on-camera actor carries in face, posture, and eyes has to arrive through sound alone — the catch in a breath, the smile you can hear, the weight of a shrug nobody will witness. The craft converts written copy into a living intention, so that a thirty-second car spot feels like a friend leaning over a fence, a dragon sounds like it could actually eat you, and an audiobook narrator becomes the only voice in a stranger's head for fourteen hours. The discipline is acting with a single instrument that also has to survive a long career of abuse. - heading: Core Mission markdown: >- Deliver, take after take and session after session, a performance that lives entirely in sound — true to the character and the client's intention, clean enough to edit, and repeatable weeks later — without wrecking the only voice you'll ever get. - heading: Primary Responsibilities markdown: >- The visible work is talking into a microphone. The actual work is reading intention off a page and committing to it instantly. A voice actor analyzes copy and finds the want underneath the words; cold-reads scripts they've never seen and lifts them off the page so nothing sounds read; matches an existing performance in ADR and dubbing, hitting the flap and the timecode; builds and guards a bible of distinct character voices; takes direction fast in a booth that costs money by the minute, and self-directs alone in a closet at home; manages breath, stamina, and vocal health across screaming animation days and marathon audiobook sessions; keeps performances consistent across recordings made days or weeks apart; and runs a small business — the demo reel, the agent, the home rig, the invoices, the rate card. Under all of it: protect the instrument, because no warmup fixes a blown-out voice on a shoot day. - heading: Guiding Principles markdown: >- - **Acting first, voice second.** A pretty voice with no intention is wallpaper. Every line answers a question: who am I talking to, and what do I want from them right now? - **Lift it off the page.** The audience must never hear you reading. The performance lives in the gaps the punctuation didn't mark. - **Serve the copy and the client.** Your job is their intention, not your showcase. The read that sells the product beats the read that flatters you. - **Protect the instrument.** Hydration, warmups, and knowing when to stop are not vanity — they're how the career lasts past forty. - **Clean is kind.** Mouth clicks, plosives, and inconsistent levels become the editor's misery and your reputation. Record like someone has to cut it. - **Commit fully, then take the adjustment.** Give a real choice on take one; then change it completely without ego when the director redirects. - **Range is a tool, not a trick.** A second voice is worthless if you can't reproduce it in three months for the pickup session. - heading: Mental Models markdown: >- - **The voice as a body.** Physicality you can't see still has to be there. If the character is running, you stand up and move; if he's exhausted, you let the breath go ragged. The mic hears the body even when the camera can't. - **Proximity as performance.** Distance to the mic is an acting choice, not just a level. Close and quiet is intimate; backing off opens a room. The proximity effect — bass buildup up close — is a color you spend or avoid. - **Intention over inflection.** Don't decide how a line sounds; decide what the character is doing to the listener — warning, seducing, pleading — and the sound follows. Actors who play sound make "announcer voice." - **The flap is the law (in dubbing).** In ADR and lip-sync the picture is fixed. The mouth movements — the flaps — and the timecode are constraints you perform inside, like writing to a fixed meter. - **Conversational baseline.** Most modern commercial and narration reads aim for "talking to one person," not broadcasting to thousands. Picture a single listener and shrink the room. - **The instrument has a budget.** Screaming, growls, and rasp spend a finite daily allowance. Bank the violent stuff for the end of the session or you'll pay for it on lines that matter. - **Consistency as continuity.** A character recorded across many sessions is one performance stitched together. Reference files, level notes, and your own muscle memory are the stitching. - heading: First Principles markdown: >- - The listener builds the picture; you supply only the sound that makes them build the right one. - Anything that breaks the illusion — a click, a pop, a read that sounds read — is a defect, however beautiful the voice around it. - The microphone is unforgiving and literal: it captures exactly what you do, including the breath you thought you hid. - A voice has finite mileage per day and per lifetime; you are spending a non-renewable thing. - heading: Questions Experts Constantly Ask markdown: |- - Who am I talking to, and what do I want from them in this line? - What's the spec really asking for under the adjectives — "warm and authoritative" means what, exactly? - Where's the operative word in this sentence? - Am I matching the scratch track / the existing character, or creating? - Will this take cut cleanly, or did I click, pop, or breathe into it? - Can I do this voice again in three months without hurting myself? - Is the read selling, or is it just sounding nice? - How much voice do I have left today, and what still has to be screamed? - heading: Decision Frameworks markdown: >- - **Spec decode.** Translate vague direction ("more genuine," "less salesy," "younger") into a concrete acting adjustment — a person, a relationship, a stake — before the next take. Never just "do it again, better." - **Create vs. match.** If there's a scratch track, a previous actor, or approved earlier sessions, your job is fidelity; suppress invention. If it's an original, you own the choice — bring two distinct options. - **Proximity and angle per moment.** Decide mic distance and off-axis angle by the intention and the plosive risk of the line, not once at the top. - **Stamina triage.** Sequence a session so fragile, nuanced lines are recorded fresh and the throat-shredding work goes last. Ask to reorder if the engineer scheduled it backward. - **Buyout vs. royalty math.** Weigh a flat buyout against likely usage. A national broadcast spot with residuals is worth holding out for; a one-day e-learning module is a buyout and you move on. - heading: Workflow markdown: >- 1. **Receive.** Script and spec land — sometimes the night before, often on arrival. Note usage, length, character notes, pronunciations, and timecode if it's to-picture. 2. **Cold read / prep.** Read aloud once for sense, mark operative words, breaths, and the trap words (numbers, names, tongue-twisters). On a true cold read you do this in your head while the red light is already coming up. 3. **Warm up.** Lip trills, sirens, tongue twisters, hydrate. Never the first take of the day on a cold voice. 4. **Set the chain.** Mic at the right distance and angle, pop filter placed, headphones balanced so you hear yourself honestly, room tone respected — kill the fridge and the HVAC. 5. **Take one, committed.** Give a full, specific choice. Slate if asked. 6. **Take direction.** Adjust fast and completely; bank a safety take and an alt. In ADR, watch the flap and the streamer, hit the in and out. 7. **Cover yourself.** Record pickups, wild lines, and alternate reads while you're warm and matched. 8. **Self-direct (home).** Without a director, you are your own ears — listen back ruthlessly, punch in clean, log levels and mic position for next session's consistency. 9. **Deliver.** Edited or raw per spec, named correctly, breaths and clicks handled, room tone bed included if requested. - heading: Common Tradeoffs markdown: >- - **Intimacy vs. plosives.** Working close gives warmth but invites pops and proximity boom; back off and the magic cools. The pop filter and an off-axis angle are the compromise. - **Big choice vs. directable range.** A huge swing reads great on take one but may be hard to dial back; leave the director somewhere to go. - **Vocal effect vs. longevity.** The rasp the client loves today can cost you the rest of the week. Fake it with placement before you grind real cords. - **Speed vs. accuracy in the booth.** Studio time is expensive; few takes. Nail it fast, but a clean third take beats a flubbed first. - **Buyout cash now vs. residual upside.** Guaranteed money against the gamble that the spot runs nationally for two years. - **Versatility vs. a signature.** Casting hires a known color, but a one-trick voice books less. Most careers balance a recognizable home base with reach. - heading: Rules of Thumb markdown: >- - Smile and they hear it; frown and they hear that too. - Find the one word the sentence is really about and lean there. - If you popped a plosive, turn slightly off-axis and try again — don't fight it. - Stand for energy; sit for confidential. - Drink room-temperature water, never iced; skip dairy before a session. - Mark your breaths so the editor doesn't have to hunt them. - When in doubt on a cold read, play it to one person, conversationally. - The screaming and the death sounds go last in the day, always. - Log your mic distance and gain; future-you re-records the pickup. - Slate clearly; the file with no slate is the file nobody can find. - heading: Failure Modes markdown: >- - **The announcer voice.** Reaching for a generic "broadcast" sound instead of talking to a person. Dated, and it tells the listener nothing. - **Reading, not acting.** Honoring the punctuation and stress patterns of text rather than the intention; every line lands the same shape. - **Wrecking the voice.** Screaming or rasping without support and being hoarse by lunch, or for a week. - **Inconsistency across sessions.** A character whose pitch, energy, or accent drifts because no reference was kept. - **Popping and clicking through it.** Beautiful performance, unusable file; dry mouth and bad mic discipline make an editor hate you. - **Indirectable.** Giving one fixed read and being unable to adjust when the director asks for something new. - **Over-narrating an audiobook.** Acting every character so hard the prose drowns; the author's voice disappears under yours. - heading: Anti-patterns markdown: >- - **"I'll fix it in my head" cold reads** — skipping the pronunciation check and fumbling the brand name on the master take. - **Set-and-forget proximity** — locking one mic distance for an entire scene regardless of whisper or shout. - **Ego on the adjustment** — defending your choice instead of taking the note. - **Self-directing by vibe** — at home, not listening back, shipping clicks and inconsistent levels. - **Demo reel as résumé dump** — a reel of everything you can sort-of do instead of the three things you book. - **Working sick or unwarmed** — and blaming the cold for a performance you could have protected. - heading: Vocabulary markdown: >- - **The copy** — the script, especially in commercial work. - **The spec** — the client's brief on tone, audience, and read style. - **Cold read** — performing a script seen for the first time, often on the spot. - **Operative word** — the word a line hinges on; where the meaning lives. - **Proximity effect** — bass boost when you work very close to the mic. - **Plosives / sibilance** — popping "p"/"b" bursts and harsh "s" hiss. - **Pop filter** — the screen that tames plosives. - **ADR / looping** — re-recording dialogue to picture in post. - **The flap / lip-flap** — the on-screen mouth movements you must match. - **Scratch track** — a rough reference recording you perform against. - **Room tone** — the recorded silence of the space, used to fill edits. - **Wild lines** — lines recorded off-picture, free of timecode. - **The pickup** — a re-recorded line or fix, often a later session. - **Buyout** — a flat fee that signs away usage rights, no residuals. - **Rate card** — your published prices by usage and length. - **Slate** — announcing take/name/file at the head of a recording. - heading: Tools markdown: >- - **The microphone** — large-diaphragm condenser in studio; a known, neutral mic at home. You learn its sweet spot and its temper. - **The pop filter and the booth** — to tame plosives and kill the room. - **The home studio / DAW** — a treated closet, an interface, and software to record, punch in, edit breaths and clicks, and deliver. - **Headphones** — closed-back, to monitor honestly without bleed. - **The script and timecode/streamer** — your map; in ADR the streamer or beeps cue the in-point. - **The demo reel** — the single most important sales asset; short, current, and genre-specific. - **Water, steam, and warmups** — unglamorous, load-bearing. - heading: Collaboration markdown: >- Voice work is solitary at the mic and deeply collaborative around it. The director owns the intention and the casting; the engineer owns the signal, levels, and the clean file; the producer and client own usage, brand, and budget. In animation and games you often record alone, months before or after your scene partners, so the director carries the continuity you can't hear. In ADR and dubbing you serve the editor and the picture. Audiobook narration adds the author (sometimes) and the proof-listener who catches every mispronunciation and mouth click. The friction lives at the seams: a spec that says "edgy but safe," a scratch track you can't match, a character recorded by a different actor last season. The professional over-asks at exactly those seams and never makes the engineer guess. - heading: Ethics markdown: >- Your voice is your identity and your livelihood, which makes consent and ownership the central ethical questions. Read the contract: a buyout that quietly includes "synthetic and AI uses in perpetuity" can train a model to replace you with your own voice. Don't lend your instrument to deception you'd be ashamed of — fake testimonials, scams, deepfaked words you never said. Respect the chain: don't undercut scale rates to the point of breaking the market for everyone, and don't sound-alike another actor's signature role to dodge paying them. Disclose health limits honestly rather than blowing a cord to hit a deadline. AI voice cloning sits over the whole craft now — sometimes a legitimate tool with informed consent and fair pay, sometimes theft. Knowing which is which, and refusing the theft, is part of the job. - heading: Scenarios markdown: >- **A national commercial cold read.** You walk in, the spec says "confident but approachable, like a friend who happens to be an expert." First instinct is not to perform but to decode: a friend, not an announcer; one listener, not a crowd. You scan for the operative words and the trap — the drug name nobody can pronounce — and check it before the red light. Take one: warm, close to the mic, talking to a single person across a kitchen table. The director says "less sell." You don't just soften the volume; you drop the stakes — now you're mentioning it, not pitching it — and back off the mic an inch so the room opens. Three takes, a safety, an alt with a smile in it. Done in ten minutes because the booth costs by the minute and you read the intention, not the words. **An animation screaming day.** The session has forty lines, including a character falling off a cliff and a two-minute battle of efforts and death sounds. You see the schedule puts the screams first and ask to reorder — fragile emotional lines while the voice is fresh, the throat-shredding work last. You warm up properly, support every scream from the breath rather than the cords, and place the rasp forward in the mask so it sounds violent without grinding the folds raw. The director wants the death sound "bigger." You give it physically — stand, drop your weight, throw the sound — and bank exactly one more take, because there isn't a third in you today and pretending there is means a week hoarse and a blown pickup session. **Matching yourself months later in ADR.** A pickup comes in: one line for a character you voiced in a different studio eleven weeks ago. The picture is locked, so the flap and the timecode are fixed. Before you open your mouth you pull your reference file and your logged mic distance and gain, match the energy and pitch to the old performance, and play the line into the existing scene, not into a vacuum. You watch the streamer, hit the in-point, land inside the flap, and record three matched passes plus a wild line in case the sync is tight. The test isn't whether this take is good — it's whether anyone can hear the seam. They can't. - heading: Related Occupations markdown: >- A voice actor shares the actor's core craft — intention, given circumstances, truthful choices under direction — but works without face or body, building everything from sound. The sound engineer is the constant partner who turns the performance into a clean, usable signal and lives in the same vocabulary of proximity, plosives, and room tone. Animators give the character its body and timing that the voice must match; the film editor and the dubbing editor cut the takes and depend on clean breaths and consistent levels. Musicians share the discipline of an instrument that needs warmup, breath support, and lifelong care. Broadcast journalists work the same microphone and conversational baseline, though toward truth rather than character. - heading: References markdown: |- - *Word of Mouth* — Susan Blu & Molly Ann Mullin - *There's Money Where Your Mouth Is* — Elaine Clark - *The Art of Voice Acting* — James Alburger - *Voice-Over Voice Actor* — Yuri Lowenthal & Tara Platt