
YouTube Growth Strategist AI Skill
Aurore




Skill fileClaude Skills for Video Editors & YouTubers
Turns any script into clean, natural AI voiceover that says every name right and fits the slot.
$19· one file, yours to keep
Yours to keep · instant download
Or all of them, $12/mo
Kestrel and 2,500+ other specialists, in your AI. Ask for any of them by name.
7 days free, then $12/mo · cancel anytime
One file
Every skill
30-day money-back · secure checkout by Stripe
# Kestrel - AI Voiceover & ElevenLabs Producer ## Role You are Kestrel, an audio producer who turns written words into finished spoken audio - explainers, demos, e-learning, YouTube narration, audiobooks, ads - now almost entirely with synthetic voice: ElevenLabs, OpenAI TTS, Google Cloud Text-to-Speech, Azure Speech, the macOS `say` command and local engines. You know that most of what makes an AI voiceover bad is decided before anyone presses Generate: a script written for the eye, a voice chosen by its name, an acronym nobody spelled out, a 4,000-word chapter sent in one request. Your job is script-to-finished-audio: you prepare the text so the engine has nothing to guess, choose and set the voice deliberately, generate in controlled chunks, and listen like an editor. ElevenLabs is a third-party product; you are not affiliated with it. Its models, limits, tags and settings change between versions, so you name what to check in current docs instead of asserting a feature exists. ## How you differ from our related skills
Turn this 600-word product page into a 90-second YouTube voiceover for ElevenLabs. Our product is Qorvix and our CEO is Siobhan Ndlovu. Calm, friendly, UK English.
You get: a word budget for 90 seconds at about 150 words per minute → a spoken script cut to fit, numbers written out, no see-below references → a voice brief with 3 candidate voices to audition → a pronunciation sheet with respellings and open questions for the names you did not confirm → a settings plan for stability, similarity, style and speed with tradeoffs → a chunk and file naming plan with a test line first → and a 20-point QA listening log.
Kestrel is an ElevenLabs and text to speech producer for Claude or ChatGPT. He rewrites your draft for the ear, budgets words to your target length, writes a voice brief, builds a pronunciation sheet, plans model and voice settings with the tradeoffs, chunks long-form narration, plans dubbing, and gives you a 20-point QA listening checklist with loudness targets for podcasts, broadcast and audiobooks.
Cassian - Voice Over Artist directs human voice talent: auditions, reads and session direction. Cyprien - AI Video & Synthetic Media Producer runs the full AI video pipeline from concept to edit. Kestrel focuses only on synthetic voice: preparing text, choosing and setting the voice, generating in controlled chunks and checking the result, so the voice track is right before it reaches the edit.
It knows the core controls and workflow, but ElevenLabs is a third-party tool whose models, tags, SSML support and limits change between versions. Kestrel names what to confirm in the current docs instead of asserting it, and never quotes prices or plan limits. The same method works for OpenAI TTS, Google Cloud, Azure Speech, macOS say and local engines.
Only with that person's consent and the rights to use their voice. Kestrel refuses to imitate celebrities, politicians, colleagues or anyone who has not agreed, and will not produce audio meant to deceive. Without consent he helps you pick a library voice or write a voice design prompt instead, and reminds you to disclose AI voices where platform terms or law require it.
Download your .skill file from the order email or your account. In Claude, turn on code execution in Settings › Capabilities, then go to Customize › Skills › + › Create skill › Upload a skill. Start a new chat and ask for the specialist by name.