▸ status log

VoxCPM Voice Cloning

Zero-shot voice synthesis · pinkie RTX 3060
checking… v1.0
Enroll from a recording
Cleaned on bigfoot (DeepFilterNet), then split by speaker and transcribed (Parakeet). ~1 min for a 15 min file.
Run step A first.
Names from step B become voices. Mode: zero-shot is instant. They appear in the Voice list for synthesis.
Your new voice is selected in Synthesize above — type text and Generate.
Synthesize
Documents
Drop files here — or click to choose
no files selected
Documents live on bigfoot and show up in the iPhone app's library under “On bigfoot”. PDFs keep their first 500 pages.
People
A render is hours of the one GPU, so anyone who is not an admin has their submissions held here until you say yes to that particular job.
Each person gets a token. The app and this page send it with every request; anyone without one is public and sees only what has been shared with everyone. While there are no accounts at all, every request is treated as an admin — creating the first one is what turns the lock on, so make it your own and sign in with it before handing anyone else a token.
Storage
loading…
Re-encodes every archived sentence from wav to FLAC — lossless, and verified sample-for-sample before the wav is deleted. Gives back about 58%. The archive keeps working exactly as before: playback still hits it, audiobooks still assemble from it, and a cast change still re-renders only the lines that changed. A book being rendered right now is skipped.
One-off Synthesize outputs and finished reader jobs, older than the given age. Render archives are never touched by this.
days
Reader
Default voice: Pemberton. Paste long text and the reader will queue chunks for continuous playback.
Loads a document from the library above straight into the text box.
On iPhone, tap this to take a picture of a printed page and have it read aloud.
Create a voice — record
loading…
Letters, digits, _, -. No spaces. Must start with a letter.
Create a voice — upload
Letters, digits, _, -. No spaces.
Create a voice — design

Describe a voice and the model invents one — no reference recording, no licence to check. Each take is a different person, so roll until you hear one you want, then keep it. Keeping freezes that take as the reference clip, after which it behaves like any other enrolled voice.

Gender, age, accent, tone, pace. Concrete beats poetic.
Manage voices
loading…
Train LoRA — improve a voice

LoRA fine-tuning teaches the model a specific voice more deeply than zero-shot cloning. Use when zero-shot output doesn't capture cadence, accent, or character well enough.

How it works:

  1. Upload 5–30 min of clean, single-speaker audio.
  2. Provide a transcript. Either attach one you already have (.txt, .srt, .vtt, .json) or click Transcribe with Whisper. Whisper is only started if no transcript is supplied.
  3. Review the segments; edit any misheard text in place.
  4. Name the new voice (e.g. Bradley_LoRA1) and submit. Training runs in the background on pinkie's 3060 (~15–60 min).
  5. When done, the new voice appears in the list with a [lora] marker. Select it and synthesize as usual — the adapter hot-swaps automatically.

Faster path: to train on the RTX 3090 and keep synthesis online, use the Enrollment Wizard above (Step B → per-speaker LoRA mode) — it trains on bigfoot with audio-aligned Parakeet transcripts. This upload flow is the manual, single-file alternative.

Note: this upload flow trains on pinkie's 3060 and pauses synthesis until done (VoxCPM auto-reloads on your next Generate). Transcription (step 2) now runs on bigfoot's 3090, so only the training step is local.

Accepted: plain .txt (aligned via silence detection), .srt/.vtt with timestamps, or .json in the form {"segments":[{"start":0,"end":5,"text":"..."}]}. Leave empty to transcribe with Whisper in the next step.