Hallo

Paid LinuxSelf-hostedWindows

What Hallo is

Hallo turns a still portrait and speech audio into an animated MP4 video. Its hierarchical audio-driven synthesis is designed to coordinate lip movement, facial expression, and pose with the audio. During inference, separate weights let you adjust pose, face, and lip motion. The source image should be square, with a forward-facing face filling 50%–70% of the image and rotated less than 30 degrees. The driving audio must be a clear-vocal English WAV file; background music is acceptable. The video is saved to `.cache/output.mp4` by default, with an option to choose another output path.

The listed system requirements are Ubuntu 20.04 or 22.04 and CUDA 12.1; the project says it tested the software on A100 GPUs. Pretrained Hallo checkpoints and supporting models are provided through its Hugging Face repository. The project is free and open source under an MIT license, and lists Linux, Windows, and self-hosted platforms. Its repository links community resources for Windows, ComfyUI, a WebUI, Docker, and RunPod. Training and data-preprocessing code is also available, including support for multi-node and multi-GPU training. The training datasets are English, so driving audio must be in English; the roadmap mentions improving Mandarin performance. The project also identifies deepfake misuse and image and voice privacy and consent as social risks, and calls for responsible use and privacy safeguards.

Who it suits

Hallo suits people who want to animate a portrait from a still image and English speech audio, and who can work with its listed system requirements. It also offers training code for users working with their own training setup.

What it does well

Free and open source under an MIT license.
Separate inference weights control pose, face, and lip motion.
Saves the generated animation as an MP4 video.
Provides pretrained checkpoints and supporting models.
Includes data-preprocessing and multi-GPU training code.

What to know first

Driving audio must be an English WAV file.
Listed system requirements are Ubuntu 20.04 or 22.04 and CUDA 12.1.
The source portrait must meet specific framing and angle requirements.

Limits

Driving audio must be English, and the source image must be square with a forward-facing face occupying 50%–70% of the frame and rotated less than 30 degrees. The listed system requirements are Ubuntu 20.04 or 22.04 and CUDA 12.1.

Getting started

Open the Hallo project website.
Use Ubuntu 20.04 or 22.04 with CUDA 12.1.
Get pretrained Hallo checkpoints and supporting models from the project's Hugging Face repository.
Prepare a square, forward-facing portrait with the face occupying 50%–70% of the image and less than 30 degrees of rotation.
Prepare clear-vocal English audio as a WAV file.
Run inference with the source image and audio; choose an output path or use the default `.cache/output.mp4`.

Verdict

Choose Hallo if you need to generate an animated portrait from a still image and English audio, and its system requirements fit your setup. Look elsewhere if you need driving audio in another language or cannot provide a suitably framed portrait.

Questions

Is Hallo free?

Yes. Hallo is listed as free and open source.

What license does Hallo use?

The GitHub repository lists an MIT license.

Which platforms does Hallo support?

The listed platforms are Linux, Windows, and self-hosted. The repository links a community Windows version and installation guide.

What audio and image can I use?

The driving audio must be a clear-vocal English WAV file; background music is acceptable. The source image should be square, forward-facing, and meet the stated face size and rotation requirements.

What does Hallo produce?

It generates an animated portrait video and saves an MP4 by default to `.cache/output.mp4`; you can choose another output path.

Who makes Hallo?

The listed authors are affiliated with Fudan University, Baidu, ETH Zurich, and Nanjing University.

Facts from Hallo's own pages

Audio requirements: Driving audio must be a WAV file in English, with clear vocals; background music is acceptable. · github.com, 8 Oct 2026
Community integrations: The repository links community resources for ComfyUI, a WebUI, Windows, Docker, and RunPod. · github.com, 4 Oct 2026
Founded: 2024 · fudan-generative-vision.github.io, 28 Sept 2026
Image requirements: The source image should be square, show the face as 50%–70% of the image, and have a forward-facing head with less than 30° rotation. · github.com, 8 Oct 2026
Input image requirements: The source image should be square, with a forward-facing face occupying 50%–70% of the image and rotated less than 30 degrees. · github.com, 4 Oct 2026
Inputs and output: Inference takes a source image and driving audio and saves an animation as an MP4 video by default. · github.com, 8 Oct 2026
Installation: The project lists Ubuntu 20.04 or 22.04 and CUDA 12.1 as system requirements, and says it tested the software on A100 GPUs. · github.com, 8 Oct 2026
Language limitation: The README says the model's training datasets are only in English and lists improving Mandarin performance as a roadmap item with no estimated date. · github.com, 8 Oct 2026
License: GitHub identifies the repository as MIT-licensed. · github.com, 8 Oct 2026
Model downloads: Pretrained models for inference are available from the project's Hugging Face repository, with separate downloads also listed for components such as wav2vec and Stable Diffusion 1.5. · github.com, 8 Oct 2026
Motion controls: Inference options include separate pose, face, and lip weights. · github.com, 4 Oct 2026
Motion synthesis: Its hierarchical audio-driven synthesis is designed to align lip, expression, and pose motion with audio. · arxiv.org, 4 Oct 2026
Output: Inference saves an MP4 video to `.cache/output.mp4` by default, with an option to choose another output path. · github.com, 4 Oct 2026
Privacy and misuse: The project identifies deepfake misuse and image and voice privacy and consent as social risks, and calls for responsible use, informed consent, and privacy safeguards. · github.com, 4 Oct 2026
Purpose: Hallo is a system for audio-driven visual synthesis that animates portrait images. · github.com, 8 Oct 2026
Research team: The listed authors are affiliated with Fudan University, Baidu, ETH Zurich, and Nanjing University. · github.com, 8 Oct 2026
Responsible use: The project notes risks involving deepfakes, privacy, and consent, and calls for ethical guidelines, transparent data use, informed consent, and privacy safeguards. · github.com, 8 Oct 2026
System requirements: The listed system requirements are Ubuntu 20.04 or 22.04 and CUDA 12.1; the tested GPU is A100. · github.com, 4 Oct 2026
Training: The project provides data-preprocessing and training code, including support for multi-node and multi-GPU training. · github.com, 8 Oct 2026
Windows and integrations: The project README lists community contributions for a Windows version, ComfyUI integration, WebUI, and a Docker template. · github.com, 8 Oct 2026
Windows support: The repository links a community Windows version and installation guide. · github.com, 4 Oct 2026

Where Hallo stands on Freedom251

AI Photo Animation Tools#20 of 25
AI Video Lip Sync Tools#23 of 28