Train a custom voice

Last updated: November 3, 2026

Clone a voice from a sample, design one from a description, or bring in an ElevenLabs voice by its ID, so a character always sounds the same.

A trained voice lives in a reference set, next to the character's art, so the voice and the look travel together. Train it once and every speech generation that uses the set reads its script in that voice.

Before you start

  • Training needs the Train styles permission in your workspace role.

  • To clone, put one or more clean recordings of the voice in a reference set: one speaker, no music, no background noise. Voice models differ in how many clips they use and how long a clip can be; see Trainable models. If the set doesn't exist yet, create a reference set first, or open Clone voice from the side panel's tools, which starts a new one.

  • To design a voice instead, you need no clips, only a description of it.

  • To bring in a voice that already exists on ElevenLabs, you need no clips, only its voice ID. See Bring in a voice you already use on ElevenLabs below.

  • Training costs Creative Units; the price shows on the Train voice button. See Creative Units.

Steps

  1. Open the reference set that holds your sample. On a session board, double-click its tile.

  2. In the Audio section, select Design voice. The setup opens on a voice model under Base model; open the picker to choose a different one.

  3. Under Method, choose Clone to copy the voice from your clips, or Design to build one from a Voice description (tone, age, accent, delivery). Some voice models ask for more: a Reference transcript of the exact words in your clip, which improves the clone, or Clip to clone when the set holds several clips and the model clones from one. ElevenLabs models also offer Voice ID, described below.

  4. Select Train voice. The voice appears under Audio marked Training. It trains on its own, so you can keep working; you get a notification when it is done.

  5. When the row reads Ready to use, open it and select the Finetunes tab. Play the audition clip to hear the voice before you use it.

  6. Back on the board, open Create speech, pick a voice model your voice works on, then select Reference set and add your set. It lands in the prompt box as a chip. A voice trained on an ElevenLabs model works on every ElevenLabs voice model; any other voice works on the model it trained on.

  7. Type the script exactly as it should be spoken and select Create speech.

Bring in a voice you already use on ElevenLabs

When a character already has a voice on ElevenLabs, bring it into a reference set by its voice ID instead of cloning it again. You can bring in any voice from the ElevenLabs voice library, and any voice made on Layer. A voice saved only to your own ElevenLabs account can't be reached from Layer. Copy the ID from the voice's page in ElevenLabs.

  1. In the side panel's tools, search for Clone voice and select it. A new reference set opens on its training setup. To add the voice to a set you already have instead, open the set and select Design voice in its Audio section.

  2. Under Base model, pick an ElevenLabs voice model.

  3. Under Method, choose Voice ID and paste the ID into the Voice ID field.

  4. Select Train voice. Layer speaks one test line in the voice to check that it can use it; nothing else is trained.

  5. When the row reads Ready to use, open it and select the Finetunes tab to play the test line. Use the voice in Create speech as in steps 6 and 7 above.

Result

The line lands on the board as an audio file, read in your trained voice. The voice itself stays in the reference set: under Audio in its Finetunes column, with every version on the Finetunes tab. Anyone in the workspace who can use the set can voice lines with it. A voice trained on an ElevenLabs model speaks on every ElevenLabs voice model, so you can switch models without training it again.

Production tips

  • Record the sample for the job. A clone copies what it hears: the room, the microphone and the delivery. Record in a quiet room, at the energy level the character usually speaks at, and leave out music and effects. A minute or two of varied sentences gives a steadier voice than one dramatic line.

  • Keep one voice per set, next to its character. Put the voice in the character's own reference set, so a single set carries both the look and the voice into every generation.

  • Bring in a cast you already have. When your characters are already voiced on ElevenLabs, bring each voice in by its ID rather than cloning it from exported audio. The voice stays exactly the one your team signed off on.

  • Design a voice you can't record. For a creature, a narrator you haven't cast yet or a placeholder for scratch audio, Design gives you a consistent voice from a description. Describe it the way you would brief an actor: age, timbre, accent, pace.

  • Write direction outside the script. The prompt is what gets said. Where the model offers delivery controls under Advanced in Create speech, put direction there, as Create speech and voice-over describes.

  • Retrain without breaking lines in production. Retraining adds a new version; the current one keeps working until the new one is ready, and you can switch back from the Finetunes tab.

  • Carry the voice into video. Use the finished line to lip sync a character to a voice line.

If something goes wrong

  • Train voice is greyed out. Hover it for the reason: the set has no clip to clone yet, a clip is still processing, the model clones from one clip and you haven't chosen which, Design has no description, or Voice ID has no ID (an ID is letters and digits only). Design on some models takes no reference clips at all; remove the audio from the set or choose Clone.

  • Training failed with a message about the voice ID. ElevenLabs refused the test line for that ID. Check that you copied the whole ID and that the voice is in the ElevenLabs voice library or was made on Layer. A voice saved only to your own ElevenLabs account can't be used; clone it from a recording instead.

  • Training failed because of a sample. A message about length means the clip is longer or shorter than the voice model accepts: trim it to a few minutes of clean speech, or use a longer clip (limits per model are in Trainable models). A flagged sample means the provider's content policy rejected it: remove it. Then train again. For anything else, see Training failed.

  • Your voice doesn't appear in the Create speech picker. The picker shows only sets with a voice for the model in the form. Pick a model your voice works on first, then add the set. A voice trained before a model was added to Layer doesn't reach that model; select Retrain voice on it to add it.

Related