Features Pricing
Blog Documentation Changelog API Reference
Sign In
Get Started

Voice cloning

4 min read

Use a voice you have cloned in your ElevenLabs account for a consistent brand sound across every audio file. Your cloned voice works just like any other TTSWP voice. Pro and Agency plans.

Clone a voice in ElevenLabs

Voice cloning happens in your ElevenLabs account, not in TTSWP. Steps:

  1. Sign in to elevenlabs.io.
  2. Go to Voices → Create voice → Instant Voice Clone (or Professional Voice Clone for higher quality).
  3. Upload 1-3 minutes of clean voice recordings (for Instant) or 30+ minutes (for Professional).
  4. Let ElevenLabs process the audio.
  5. When ready, your cloned voice appears in your ElevenLabs voice library with a unique voice ID.

Copy the voice ID - you will need it in the next step.

Connect the cloned voice to TTSWP

Option A: via your ElevenLabs key (BYOK)

If you have your ElevenLabs API key connected to TTSWP:

  1. Go to Text to Speech → Voices.
  2. Click Refresh voices in the topbar.
  3. Your cloned voice now appears in the voice list.
  4. Click Select on it.

That is it. Cloned voices show with a "Cloned" badge.

Option B: paste the voice ID into a shortcode

You do not need to set it as default. Use it only in specific shortcodes:

[mementor-tts voice="your-cloned-voice-id"]
This post uses my cloned voice.
[/mementor-tts]

Requirements

  • An ElevenLabs account with access to voice cloning (Instant requires Starter plan; Professional requires Creator plan or higher)
  • Your ElevenLabs key connected to TTSWP via BYOK

Voice cloning on ElevenLabs is a paid feature of their platform, separate from your TTSWP subscription.

Clone your own voice only

ElevenLabs requires explicit consent to clone any voice. You can clone:

  • Your own voice
  • A voice you have written legal permission to use

You cannot clone:

  • Celebrity voices without consent
  • Voices from public figures without a licensing agreement
  • Voices from any person without documented permission

TTSWP does not police this - ElevenLabs does on their platform. But using unauthorized cloned voices exposes you to legal risk.

Best practices

Record high-quality source audio

For Instant cloning:

  • Quiet room, no echo
  • Consistent volume
  • Clear diction, natural pace
  • 1-3 minutes of material

For Professional cloning:

  • 30+ minutes of clean studio-quality recording
  • Same microphone, same environment for all samples
  • Varied emotional tones if you want expressive range

Test before rolling out site-wide

Clone a voice and test it on one post first. Some voices clone better than others. If the clone sounds off, re-record your source audio and try again.

Use the right model with cloned voices

  • Eleven Flash v2.5: works with Instant clones
  • Multilingual v2: best pairing with Professional clones
  • Eleven v3: most expressive output, works with either clone type