ElevenLabs Prompts for Voiceover, Voice Design & Dubbing

<context> I need a voiceover script for an explainer video about [PRODUCT OR TOPIC]. Audience: [TARGET AUDIENCE]. Tone: friendly, clear, conversational. Target length: [DURATION, e.g. 60 seconds] which is roughly 150 words per minute. </context> <task> 1. Write the script as natural spoken language, not written prose, using short sentences and contractions. 2. Open with a hook in the first 5 seconds that names the problem the viewer feels. 3. Spell out numbers, acronyms, and symbols phonetically so the TTS reads them correctly (e.g. write twenty four seven, not 24/7). 4. Break the script into short paragraphs so I can paste each into ElevenLabs as a separate generation. 5. End with one clear call to action. Return only [YOUR SCRIPT], no stage directions. </task>

<context> Rewrite my draft into a YouTube narration voiceover script. Draft: [YOUR SCRIPT] Channel style: [e.g. energetic tech reviews]. Speaking pace: brisk. </context> <task> 1. Convert any written-style phrasing into spoken-style phrasing a creator would actually say out loud. 2. Add natural connective beats between sections (e.g. so here is the thing, but watch this) without sounding scripted. 3. Keep sentences under 20 words so the synthesized voice keeps its rhythm. 4. Mark suggested pause points with an ellipsis where a breath improves pacing. 5. Flag any tongue-twister or hard-to-pronounce word and rewrite it. Return the final narration script only. </task>

<context> Write a [DURATION, e.g. 30 second] radio or pre-roll ad read for [BRAND]. Offer: [OFFER]. Vibe: [e.g. premium and reassuring]. Must mention the brand name twice. </context> <task> 1. Open with a benefit-led hook, not the brand name. 2. Write at the natural pace of a professional voice actor: roughly [WORD COUNT for duration] words. 3. Place the brand name early and once more in the closing line. 4. End with a memorable, easy-to-say tagline and the call to action. 5. Spell out the URL or phone number phonetically (e.g. brand dot com). Return the ad read with an approximate spoken runtime estimate. </task>

<context> Write the voice prompts for a phone IVR menu for [BUSINESS]. Menu options: [LIST OPTIONS]. Tone: calm, professional, patient. </context> <task> 1. Write a warm greeting, then the main menu with each option phrased as press one for X. 2. Keep each menu line short and unambiguous so callers can act before the next option plays. 3. Write hold, transfer, voicemail, and after-hours messages as separate clearly labeled lines. 4. Use neutral, evergreen wording with no time-sensitive references. 5. Spell out the callback number digit by digit. Return each prompt on its own line, labeled by its function. </task>

<context> Write narration for an e-learning module teaching [TOPIC] to [LEARNER LEVEL]. Lesson points: [LIST KEY POINTS]. Tone: encouraging, clear instructor. </context> <task> 1. Write narration in second person addressing the learner directly as you. 2. Chunk each concept into a short standalone paragraph mapped to one slide. 3. Define jargon the first time it appears in plain language. 4. Insert a brief recap sentence after every two concepts to reinforce retention. 5. Avoid filler so the spoken duration stays tight. Return the narration as numbered slide blocks. </task>

<context> Write a Voice Design description for ElevenLabs to generate a custom text-to-speech voice. Use case: epic movie-trailer narration. </context> <task> 1. Describe [VOICE DESCRIPTION] in one dense paragraph covering: gender, approximate age, accent, timbre, and pace. 2. Specify depth and resonance (e.g. deep, gravelly, resonant baritone with chest weight). 3. Specify delivery (e.g. slow, deliberate, dramatic pauses, building intensity). 4. Name the emotional register: ominous, grand, awe-inspiring. 5. Add a short sample line in that voice to seed the generation. Return the voice description plus one sample line. </task>

<context> Write an ElevenLabs Voice Design description for a friendly brand mascot voice for [BRAND]. Personality: [e.g. upbeat, playful, helpful]. </context> <task> 1. Describe [VOICE DESCRIPTION] including gender, age impression, accent, and energy level. 2. Specify warmth and brightness in the tone (e.g. bright, smiling, slightly higher pitch). 3. Describe pacing as lively but clear, never rushed. 4. Match the personality to the brand values: [BRAND VALUES]. 5. Provide a sample greeting line the mascot would say. Return the description and the sample line. </task>

<context> Write an ElevenLabs Voice Design description for a character voice: a world-weary 1940s noir detective for an audio drama. </context> <task> 1. Describe [VOICE DESCRIPTION]: middle-aged male, mid-Atlantic American accent, smoky low register. 2. Specify a tired, cynical, understated delivery with a slow cadence. 3. Add texture words like raspy, smoke-worn, gravel. 4. Note that the voice should sound like it is narrating a confession at 2am. 5. Include a sample monologue line in that style. Return the description and the sample line. </task>

<context> Write an ElevenLabs Voice Design description for a meditation and sleep-story guide voice. Goal: deeply relaxing, soporific. </context> <task> 1. Describe [VOICE DESCRIPTION]: soft, breathy, gender-neutral or as specified, soothing mid-low pitch. 2. Specify an extremely slow, gentle pace with long natural pauses. 3. Use words like hushed, tender, velvety, unhurried. 4. Note minimal pitch variation so it never startles a drifting listener. 5. Provide a sample opening line of a body-scan meditation. Return the description and the sample line. </task>

<context> Write an ElevenLabs Voice Design description for a customer-support agent voice for [BRAND]. Languages to support: [LIST LANGUAGES]. Tone: clear, patient, professional. </context> <task> 1. Describe [VOICE DESCRIPTION]: neutral accent that reads cleanly across the listed languages, mid pitch. 2. Specify a calm, articulate, reassuring delivery with crisp consonants for phone clarity. 3. Avoid strong regional slang or accent markers so it generalizes across locales. 4. Note steady pacing suitable for follow-along troubleshooting steps. 5. Provide a sample greeting the agent would use. Return the description and the sample line. </task>

<context> Rewrite my line for ElevenLabs v3 using inline audio tags to control emotion. Line: [YOUR SCRIPT] Target feeling per beat: [e.g. starts excited, ends nervous]. </context> <task> 1. Insert v3 audio tags in square brackets such as [excited], [nervous], [whispers], [sighs] directly before the words they affect. 2. Match each tag to the emotional beat described, transitioning naturally rather than all at once. 3. Do not over-tag: place a tag only where the emotion actually shifts. 4. Keep the original words intact; only add tags and minimal punctuation for pacing. 5. Briefly note which tags you used and why. Return the tagged line and the note. </task>

<context> Add pacing control to this script for ElevenLabs v3. Script: [YOUR SCRIPT] I want deliberate, dramatic pacing. </context> <task> 1. Insert pause and pacing tags such as [pause], [slowly], and ellipses where a beat improves delivery. 2. Use sentence breaks and line breaks strategically since v3 reads punctuation as timing. 3. Place a longer pause before any reveal or punchline. 4. Avoid stacking pauses back to back, which sounds robotic. 5. Mark where I might split this into separate generations for tighter control. Return the paced script. </task>

<context> Make this dialogue line feel human in ElevenLabs v3 by adding non-verbal cues. Line: [YOUR SCRIPT] Context: [e.g. two friends joking]. </context> <task> 1. Add appropriate non-verbal audio tags like [laughs], [chuckles], [sighs], [clears throat], or [gasps] where natural. 2. Place cues where a real speaker would react, not on every line. 3. Keep them subtle so the line stays intelligible. 4. Suggest one alternate version with lighter tagging for comparison. 5. Note any tag that may render inconsistently and needs a re-roll. Return both versions. </task>

<context> Add volume and intensity dynamics to this dramatic passage for ElevenLabs v3. Passage: [YOUR SCRIPT] Arc: builds from a whisper to a shout. </context> <task> 1. Open the quiet section with [whispers] and keep the words short and breathy. 2. Escalate through the middle with rising intensity cues. 3. Tag the climax with [shouting] or [yelling] for the loudest beat. 4. Use capitalization sparingly on emphasized words to reinforce intensity. 5. Resolve back to a calmer register at the end. Return the dynamically tagged passage. </task>

<context> Tag this multi-character script for ElevenLabs v3 so each character reads distinctly. Script with speakers labeled: [YOUR SCRIPT] </context> <task> 1. For each speaker, add a leading delivery tag describing their manner (e.g. [gruff], [cheerful], [British accent]) if using a single expressive voice. 2. Keep each character's tag set consistent every time they speak. 3. Note which lines would sound better generated with a separate dedicated voice instead. 4. Preserve speaker labels so I can route lines correctly. 5. Flag any accent tag that v3 handles unreliably. Return the tagged script. </task>