Leokani-tts 7M

Leokani-tts 7M

Speaks any line you type with a tiny 7M text-to-speech model that runs on your own CPU in the browser, for people who want local voice output.

Made by milliwattlabs on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: Leokani-tts 7M is a text-to-speech page. You type a line and press Speak, and a tiny voice model says it aloud. The voice is generated inside your browser, so your text never leaves your device. The model is milliwattlabs/leokani-tts-7M and it is built to run on any CPU.

Who it suits: People who want spoken audio from text without sending the text to a server.

Industry | Who uses it | How they use it | Example
Accessibility | People who like text read aloud | Hear typed text spoken on their own device | Listening to a short note
Software | Developers testing small speech models | Try a 7M model in a browser | Checking how it sounds on a laptop
Education | Students of speech technology | See a small TTS model run with no server | Typing a line and listening to it

If you need to use the voice in a commercial product, you can skip it, because the license allows non-commercial use only.

Where to use it:

• A web browser, on the Space's page at Hugging Face
• A browser on Windows, macOS or Linux, by opening a downloaded copy of index.html

Use case | Who | What they get back
Hear a line spoken | Anyone | Audio of the text you typed
Private speech output | Privacy-minded users | Speech made on the device with no text upload
Test a tiny model | Developers | A view of how a 7M voice model performs on a CPU

How to set it up: Online, nothing needs installing. A local copy needs a browser and a little Python for the download.

• A web browser
• Python and the pip tool, only for the download step

1. Install the Hugging Face hub package with pip.
2. Download the Space's files into a folder.
3. Move into that folder.
4. Open index.html in a web browser.

pip install -U huggingface_hub
hf download milliwattlabs/leokani-tts-7M-demo --repo-type space --local-dir leokani-tts-7M-demo
cd leokani-tts-7M-demo

How to use it: Open the Space's page, or index.html from a local copy. Type a line of text and press Speak. The voice plays in your browser.

hf download milliwattlabs/leokani-tts-7M-demo --repo-type space --local-dir leokani-tts-7M-demo

Example: You type a short sentence in the text box and press Speak. After a moment the sentence is spoken through your speakers. The text was handled inside the page, so nothing was sent out.

For newcomers and experts: Text to speech turns written words into a voice. To start, type one short sentence and press Speak.

The page runs the model through onnxruntime-web, with pronunciation supplied by espeak-ng compiled to WebAssembly. A web worker script and the leokani folder hold the model parts. With about 7 million parameters, the model aims for good quality on a plain CPU, and the licenses folder carries the third-party notices.

Type a line of text
espeak-ng turns the text into pronunciation
The 7M model runs in the browser
The speech plays on your device

Good to know:

• Limits: the voice model has about 7 million parameters, so check the sound yourself before you rely on it
• License: CC BY-NC 4.0 for the model, GPL-3.0 for the espeak-ng pronunciation and MIT for onnxruntime-web
• Cost to run: none named
• Data it reaches: your text stays on your device, and the page loads its model and runtime files from the repository
• Not for: commercial use, because the model license allows non-commercial use only

On Hugging Face, 13 November 2026

• Likes: 2
• Built with: static
• License: cc-by-nc-4.0

Links

• Space: https://huggingface.co/spaces/milliwattlabs/leokani-tts-7M-demo
• Files: https://huggingface.co/spaces/milliwattlabs/leokani-tts-7M-demo/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.