Log in
GuideBy HypernaturalVersion 6, updated

Cast a voice

hn/guide-cast-a-voice

How to choose a voice for a character: a short brief, a search of the ElevenLabs shared library for voices verified on eleven_v4, 3 picked on their metadata and auditioned at once on part of the script's copy (a narrator for a short reads the whole script instead, and the pick is the take), and the person's pick, with a performance guide for every script. Only when the person wants a voice matched to a face or a detailed character does a listening model rank a longer shortlist first. The person's own voice can be cloned with ElevenLabs, with their consent, from about 30 s of varied speech on a good microphone. Use for voice casting, choosing a TTS voice, voice selection, auditions.

SKILL.md

Give it a character's images (a sheet, a portrait, any number), written guidance (a brief, a document of personality references, a line of text), or both. Out comes a picked voice and a performance guide (tone, emotional range, delivery, audio tags, what to avoid, on-screen body language) to follow in every script featuring the character. The person picks by ear.

First question: does the character already have a voice? A voice from an earlier job (in its notes or a voice.md) is reused as is. Otherwise ask for the accent and any direction ("energetic, never slow") before the first take: both go into the brief. Keep it simple: "a British woman" is a search, 3 auditions and a pick.

Routes

RouteWhenCost and timeWhat you getLeaves
Library voiceDefault: the person names the kind of voice ("a British woman, warm")Search free; 3 auditions of one passage, run at once; minutesA real, stable voice the person heard on the script's own copyelevenlabs-voice-search, elevenlabs-tts-v4
Library voice matched to a characterOnly when the person asks for the voice to fit a face or a detailed characterAdds a shortlist of 12 and one listening call on the free previews to pick the 3The same, with the 3 chosen by ear against the imagesadds gemini-rank-voices
Clone the person's voiceThey want their own voice (or one they have the right to) speaking new linesA flat fee on Hypernatural's account, or one of their own plan's voice slots on their key; minutesA voice id used like a library voice, for every later takeelevenlabs-clone-a-voice, then elevenlabs-tts-v4
The person's own recordingsThey bring audio of the lines already spokenTheirsTheir recordings, used like any recorded trackelevenlabs-speech-to-text, guide-lipsync-video

To clone, ask first: the clone needs the person's consent that they have the right to the voice, and about 30 s of that one speaker alone on the best microphone they have, with varied intonation (elevenlabs-clone-a-voice has the recording, the call and the check). Audition the clone on part of the script's copy like any voice, and say where it doesn't sound like them. Voice design isn't covered.

One guidance document can cast several characters: a woman and a man cast from one presenter brief each got a voice that fits their face and the same personality.

The library route (one gate, a review)

Step 4 is the matched route's only; the rest is shared.

  1. Brief (the agent writes it from what the person said, and any images or guidance). The images set apparent gender, age and presence; the guidance sets personality and tone. For a plain request it is short. It holds: library filters (gender; one or two age buckets of young, middle_aged, old; accent only when the guidance or the person names one), 3–5 short search queries, the ideal voice in sound terms (pitch, timbre, texture, pace, warmth), an avoid list, the performance guide, and the audition passage: two or three sentences of the character's lines from the script, words unchanged, audio tags inline. With no script yet, the guidance's example line. Reference characters in the guidance (film roles) are mined for the qualities it asks to borrow, never for an impression of an actor.
  2. Search (elevenlabs-voice-search): every query × age bucket, plus each bucket unsearched. Keep only voices verified for eleven_v4 in the job's language: they are about 29% of the English library and sparse on the first pages, so page on (up to 5 × 100) until 30 remain.
  3. Pick 3 by metadata (the agent does it; the matched route picks 12): varied, best first; drop celebrity imitations, ASMR and single-purpose voices. Conversational, narrative and educational use cases suit an on-screen speaker; advertisement and social voices sell instead of talk.
  4. Listen, matched route only: gemini-rank-voices hears the 12 free previews with the character's images in one call; its top 3 go on.
  5. Auditions (elevenlabs-tts-v4): the 3, and only 3, read the audition passage with the brief's tags. Start all three runs at once. Library voices work by id without being added to the workspace. A short's narrator auditions on the whole script instead, and the picked audition is the take (guide-voice-a-script, A narrator for a short).
  6. Gate. review_choose per character, all in the same turn: the reference images as original, the brief (and on a later pass, the note that led to it) in brief, and the pass's voices as options in rank order, each with its audition and preview in files (and the listener's notes in facts on the matched route). The person picks one voice or writes a note. This is the voice's one approval: they heard it on the copy, so takes in that voice go ahead without asking about the voice again.
  7. Another pass (pass_b) if nothing fits: the note feeds a new brief ("warmer, less polished, a touch older"); voices auditioned before are skipped; pass letters are never reused. If the person asks to hear more of the same pass, audition the next 3.
  8. Record <character>/voice.md (voice id, TTS model eleven_v4, the performance guide), and record the pick where you keep the job's notes, so every script runs with the same arguments.

Choosing

  • The listener (matched route) is noisy: one voice went 9/10 to 4/10 between runs, and the director picked voices it ranked 3rd and 4th. It only narrows the shortlist to 3; the person hears those and decides. Don't bring it in for a plain request.
  • ElevenLabs library voices verified for v4 are reliable: the audition is how the voice sounds in every later take. Don't audition more than 3 or re-check the voice later.
  • The English library is mostly American: with no accent given, the search runs unfiltered.
  • Library professional clones are fine-tuned per model and only their owner can train them for v4. Untrained, they render but can drift from the sample (a director heard one "sound wrong"). Prefer verified voices.

Avoid

  • Inferring accent from a face: it comes from the guidance or the person.
  • Voices not verified for eleven_v4.
  • Trusting the listener over the person.
  • Asking a model whether a take "is the same voice as the preview": inconclusive (5–7/10 for everything). The person listens.

Related

  • guide-voice-a-script: performing the script with the picked voices and guides.

Workflows that use this skill

FAQs

How to choose a voice for a character: a short brief, a search of the ElevenLabs shared library for voices verified on eleven_v4, 3 picked on their metadata and auditioned at once on part of the script's copy (a narrator for a short reads the whole script instead, and the pick is the take), and the person's pick, with a performance guide for every script. Only when the person wants a voice matched to a face or a detailed character does a listening model rank a longer shortlist first. The person's own voice can be cloned with ElevenLabs, with their consent, from about 30 s of varied speech on a good microphone. Use for voice casting, choosing a TTS voice, voice selection, auditions.

Install Hypernatural Desktop, a free Mac app, and connect it to Claude Code, Codex or Claude Desktop. Then describe the video you want and your agent finds the skill itself, or paste this page's link (https://hypernatural.ai/skill/hn/guide-cast-a-voice) and it installs hn/guide-cast-a-voice and follows it.

Nothing. The skill is free and runs with tools on your own Mac.

Yes. Ask your agent to adjust it for your job, or to save its own version. You can publish your version under your own handle so anyone can install it.

Go make something cool.

Install the app, open your agent, describe the video. That's the whole onboarding.

Sticker HypernaturalSticker BannerSticker TVBoy