Voicing stops being a bottleneck
Every time you need a podcast, a radio advert, a video voiceover or an audiobook, depending on a voice actor (schedule, studio, post-production) holds up the launch. And generic synthetic voices sound like AI — listeners can tell.
Ethical voice cloning done properly sounds indistinguishable. We clone your voice (or that of a contracted voice artist), generate audio on demand and deliver a ready master. Audio production in batches, not drip by drip.
How we build it
1. Capture with explicit consent
We record 30-60 minutes of sample audio in the studio with the person whose voice is to be cloned. They sign a usage contract, commercial licence, and right of revocation. Without a signed contract, we do not clone. It is the ethical foundation of the service.
2. Model training
We use ElevenLabs Pro or PlayHT depending on the case. The model learns tone, rhythm, pauses, accent, and micro-variations. We validate it against human samples in blind tests before approval.
3. Generation pipeline
API connected to your DAW (Logic, Pro Tools, Reaper). You send the text, you receive the processed WAV. Equalisation, compression and reverb settings consistent with your sonic brand.
4. Re-recording support
If a take doesn’t work, we adjust the prompt, regenerate, validate. Three cycles included per piece. More than three, we discuss it.
What changes when this works
- Podcast production in the founder’s voice even when the founder isn’t recording that week.
- Radio adverts across 5 markets with the same tone.
- Spoken subtitles for social media videos — genuine accessibility, not an afterthought.
- Audiobooks and internal training at marginal cost.
Beyond voice
The service also includes jingle generation and sound design with Suno and Udio (with commercial licence), brand sounds (podcast intro, transitions, in-app notifications) and audio clean-up of existing recordings with iZotope RX10.
When we do NOT recommend this
- If your brand depends on the founder’s live presence (live interviews, events): cloning does not replace that, it only extends it.
- If your volume is 1-2 pieces a year: hiring a voice artist is still the more sensible option.
- If you work in environments where cloning raises ethical concerns (children’s education, mental health): we do not operate there.
Stack we use
- ElevenLabs Pro (voice cloning + Multilingual v2).
- PlayHT as an alternative when ElevenLabs is not a fit.
- Suno v4 / Udio for jingles with a commercial licence.
- iZotope RX10 for cleaning existing audio.
We start with a diagnostic session
Before quoting the full setup, we run a 90-minute session. We look at your audio file, your monthly volume and the legal contracts with the voices together, and we leave with an honest recommendation: whether this service fits where you are right now, or whether it makes more sense to start with something else.
We don’t charge for that session. If you’re interested, get in touch.