To use:
-
Install uv (if not already installed):
pip install -U uv
-
Clone and setup:
git clone https://github.com/notmax123/Zonos-Hebrew.git cd Zonos-Hebrew uv sync --extra compile -
Run the Gradio interface:
uv run gradio_interface.py
-
Or run a quick test:
uv run sample.py
The Gradio interface will be available at http://localhost:7860 with Hebrew text "שלום לכולם" pre-filled.
Zonos-v0.1 is a leading open-weight text-to-speech model trained on more than 200k hours of varied multilingual speech, delivering expressiveness and quality on par with—or even surpassing—top TTS providers.
Our model enables highly natural speech generation from text prompts when given a speaker embedding or audio prefix, and can accurately perform speech cloning when given a reference clip spanning just a few seconds. The conditioning setup also allows for fine control over speaking rate, pitch variation, audio quality, and emotions such as happiness, fear, sadness, and anger. The model outputs speech natively at 44kHz.
For more details and speech samples, check out our blog here
We also have a hosted version available at playground.zyphra.com/audio
The Hebrew-trained model weights are available on Hugging Face: notmax123/Zonos-Hebrew
To use the Hebrew model weights in your code, simply replace the model loading line:
# Instead of the original Zyphra models:
# model = Zonos.from_pretrained("Zyphra/Zonos-v0.1-transformer", device=device)
# Use the Hebrew-trained model:
model = Zonos.from_pretrained("notmax123/Zonos-Hebrew", device=device)Try the model instantly in your browser without any installation: 🤗 Zonos-Hebrew Space
The Hugging Face Space provides an easy-to-use web interface where you can:
- Input Hebrew text for speech synthesis
- Upload reference audio for voice cloning
- Adjust various parameters like speaking rate and emotions
- Download the generated audio directly
Perfect for quick experiments or sharing with others who want to try the model without technical setup!
Use the prebuilt image from Docker Hub:
docker pull maxme123/zonos-hebrew:latest
# Run with GPU and expose Gradio on port 7860
docker run --gpus all --rm -p 7860:7860 --shm-size=2g maxme123/zonos-hebrew:latest- Gradio UI: http://localhost:7860
- Default textbox text is set to "שלום לכולם".
- Optional public link:
docker run --gpus all --rm -e GRADIO_SHARE=1 -p 7860:7860 --shm-size=2g maxme123/zonos-hebrew:latestdocker build -t maxme123/zonos-hebrew:latest .
docker run --gpus all --rm -p 7860:7860 --shm-size=2g maxme123/zonos-hebrew:latestdocker compose up # uses the included docker-compose.yml
# or
docker compose up -d # run detached- Install NVIDIA drivers and nvidia-container-toolkit
- Verify GPU visibility:
nvidia-smi
docker run --gpus all --rm nvidia/cuda:12.4.1-cudnn-devel-ubuntu22.04 nvidia-smiZonos follows a straightforward architecture: text normalization and phonemization via eSpeak, followed by DAC token prediction through a transformer or hybrid backbone. An overview of the architecture can be seen below.
import torch
import torchaudio
from zonos.model import Zonos
from zonos.conditioning import make_cond_dict
from zonos.utils import DEFAULT_DEVICE as device
model = Zonos.from_pretrained("notmax123/Zonos-Hebrew", device=device)
wav, sampling_rate = torchaudio.load("assets/exampleaudio.mp3")
speaker = model.make_speaker_embedding(wav, sampling_rate)
torch.manual_seed(421)
cond_dict = make_cond_dict(text="ירושלים יום טוב", speaker=speaker, language="he")
conditioning = model.prepare_conditioning(cond_dict)
codes = model.generate(conditioning)
wavs = model.autoencoder.decode(codes).cpu()
torchaudio.save("sample.wav", wavs[0], model.autoencoder.sampling_rate)uv run gradio_interface.py
# python gradio_interface.pyThis should produce a sample.wav file in your project root directory.
For repeated sampling we highly recommend using the gradio interface instead, as the minimal example needs to load the model every time it is run.
- Zero-shot TTS with voice cloning: Input desired text and a 10-30s speaker sample to generate high quality TTS output
- Audio prefix inputs: Add text plus an audio prefix for even richer speaker matching. Audio prefixes can be used to elicit behaviours such as whispering which can otherwise be challenging to replicate when cloning from speaker embeddings
- Multilingual support: Zonos-v0.1 supports English, Japanese, Chinese, French, and German
- Audio quality and emotion control: Zonos offers fine-grained control of many aspects of the generated audio. These include speaking rate, pitch, maximum frequency, audio quality, and various emotions such as happiness, anger, sadness, and fear.
- Fast: our model runs with a real-time factor of ~2x on an RTX 4090 (i.e. generates 2 seconds of audio per 1 second of compute time)
- Gradio WebUI: Zonos comes packaged with an easy to use gradio interface to generate speech
- Simple installation and deployment: Zonos can be installed and deployed simply using the docker file packaged with our repository.
- Operating System: Linux (preferably Ubuntu 22.04/24.04), macOS
- GPU: 6GB+ VRAM, Hybrid additionally requires a 3000-series or newer Nvidia GPU
Note: Zonos can also run on CPU provided there is enough free RAM. However, this will be a lot slower than running on a dedicated GPU, and likely won't be sufficient for interactive use.
For experimental windows support check out this fork.
See also Docker Installation
Zonos depends on the eSpeak library phonemization. You can install it on Ubuntu with the following command:
apt install -y espeak-ng # For Ubuntu
# brew install espeak-ng # For MacOSWe highly recommend using a recent version of uv for installation. If you don't have uv installed, you can install it via pip: pip install -U uv.
uv sync
uv sync --extra compile # optional but needed to run the hybrid
uv pip install -e .uv pip install -e .
uv pip install -e .[compile] # optional but needed to run the hybridpip install -e .
pip install --no-build-isolation -e .[compile] # optional but needed to run the hybridFor convenience we provide a minimal example to check that the installation works:
uv run sample.py
# python sample.py