The Free, Local ElevenLabs (Clone Any Voice for $0)
ElevenLabs is incredible, and incredibly expensive. OmniVoice Studio is the open-source version that runs entirely on your own computer: voice cloning from a 3-second sample, voice design, video dubbing, even system-wide dictation, across 646 languages. Fully free, fully private, nothing sent to the cloud. The only catch is the setup, so we let Claude Code do that part. Paste one prompt and it installs everything.
What you're getting
OmniVoice Studio is a free, open-source, fully local alternative to ElevenLabs: Clone any voice from a 3-second clip, in 646 languages. Design a voice from scratch by tuning gender, age, accent, pitch, speed and emotion. Dub a YouTube link or a video file into another language. Dictate anywhere with a system-wide hotkey. Isolate vocals, separate speakers, batch-process, and more. It runs entirely on your machine (Apple Silicon, an NVIDIA/AMD GPU, or CPU), so it is private and genuinely free. No subscription, no credits, no catch.
What you need
A reasonably modern computer. It runs best on a Mac with Apple Silicon or a PC with a GPU, and it will still run on CPU, just slower. Claude Code installed (the free CLI from Anthropic). That is the tool that does the whole install for you, so you never touch the technical parts.
Step 1: Get the repo and Claude Code
Install Claude Code if you do not have it, then open it in a folder where you want the app to live. Here is the project it is going to set up:
https://github.com/debpalash/OmniVoice-StudioStep 2: Paste this and let it cook
In Claude Code, paste the prompt below. It clones the project, reads the install guide for your operating system, installs the dependencies and models, and launches the app. If something breaks, it fixes it and keeps going.
Set up OmniVoice Studio on my computer from https://github.com/debpalash/OmniVoice-Studio
Clone the repo, read the README and the install guide for my operating system, install everything it needs (the Python deps, the models, the backend and the frontend), then start it up. If anything fails, fix it and keep going until the app opens and the built-in self check passes.Step 3: Clone a voice
Once it is open, pick one of the built-in voices or drop in a 3-second clip to clone your own. Hit Use Voice, type what you want it to say, and it speaks. Voice design, dubbing and dictation all live in the same app.
The Claude trick
OmniVoice ships with an MCP, so you can connect it straight to Claude and just tell Claude what you want in plain English: generate a line in a cloned voice, dub a clip, design a new voice. Add the MCP once and Claude drives the whole studio for you.
One rule
Clone your own voice, or voices you actually have the rights to use. Same line you would not cross with any voice tool. Use it for your content, your dubs and your narration, not to impersonate real people.
Get the next one first
New prompts every week.
Free. The new drops and the tools behind them, before they hit the feed.
No spam · New issues Sunday · Unsubscribe anytime
Need it custom?
Want this built for you?
Tell me the idea and I’ll build it. An app, a tool, an automation. You don’t need to be technical.