ArchestackGitHub ↗ (opens in a new tab)

Open-source alternatives to HeyGen

pyvideotrans leads with 78, ahead of liveportrait (43) and duix-avatar (35). Updated from public GitHub data.

The highest-scoring open-source projects in Image and video. They are picks, not exact replacements, so read the weaknesses before you switch. The same picks fit if you are leaving HeyGen AI or Synthesia.

Try before you install
liveportrait live demo
Open-source alternatives to HeyGen
RankProjectScore
1pyvideotransVideo translation and AI dubbing pipeline with ASR, subtitle translation and TTSVoice19.6k starsGPL-3.078 out of 100
2liveportraitAnimates a still portrait from a driving video, with stitching and retargeting Live demo ↗ (opens in a new tab)Image and video19.4k starscustom license43 out of 100
3duix-avatarOffline talking-avatar video generator with voice cloning and lip-syncImage and video16.2k starscustom license35 out of 100

Reviews

178 out of 100

pyvideotrans

Video translation and AI dubbing pipeline with ASR, subtitle translation and TTS

19.6k starsGPL-3.0last commit Nov 2026

pyVideoTrans translates videos between languages by chaining speech recognition, subtitle translation, speech synthesis and audio-video merging. It runs as a PySide6 desktop GUI, a CLI, or a Gradio WebUI, and works with local models (Faster-Whisper, Ollama, F5-TTS) or online APIs. Pipeline stages can be paused for manual proofreading, and speaker diarization allows different dubbing voices per speaker.

Strengths

  • Full pipeline in one tool: ASR, translation, TTS and video merge
  • Mixes local engines (Faster-Whisper, Ollama) with many commercial APIs
  • Pause and edit between recognition, translation and dubbing stages
  • GUI, CLI for batch jobs, WebUI, and a prebuilt Windows exe

Weaknesses

  • GPU acceleration is NVIDIA only; AMD needs the Whisper.NET route
  • Windows GPU use requires manually installing CUDA 12.8 and cuDNN 9.11
  • WebUI is an optional extra and Docker image must be built locally
  • GPL-3.0 license; README lists no RAM or VRAM requirements
  • GPU optional
  • Docker
  • Needs FFmpeg, Python 3.10, uv, CUDA 12.8 and cuDNN 9.11 (NVIDIA GPU only)
  • Models: Faster-Whisper, WhisperX, OpenAI Whisper, Ollama, DeepSeek
  • port 7860
Documentation ↗ (opens in a new tab)Website ↗ (opens in a new tab)Repository on GitHub ↗ (opens in a new tab)
243 out of 100

liveportrait

Animates a still portrait from a driving video, with stitching and retargeting

Live demo ↗ (opens in a new tab)19.4k starscustom licenselast commit Jun 2026

Official PyTorch implementation of the LivePortrait paper. It animates a source image or video of a human, cat or dog using motion from a driving video or a precomputed .pkl motion template. It ships CLI scripts (inference.py, inference_animals.py) and a Gradio app with pose and regional editing, and pretrained weights are hosted on Hugging Face.

Strengths

  • Accepts image or video sources, including portrait video editing and image-driven mode
  • Motion templates (.pkl) speed up inference and avoid sharing the driving video
  • Gradio UI, Windows one-click installer and a Hugging Face Space demo are provided
  • Optional torch.compile gives 20-30% faster inference per the README (not on Windows or macOS)

Weaknesses

  • Animals mode needs X-Pose and a compiled CUDA op; tested only with NVIDIA GPUs
  • Apple Silicon is not supported for Animals mode and is about 20x slower than an RTX 4090
  • Needs conda, FFmpeg and a manual weight download; no Docker setup
  • License is not detected automatically (NOASSERTION); check the repo before commercial use
  • GPU optional
  • Needs conda, FFmpeg, PyTorch, Gradio, huggingface_hub, X-Pose (Animals mode)
  • Models: LivePortrait Humans model, LivePortrait Animals model, KlingTeam/LivePortrait weights on Hugging Face
Live demo ↗ (opens in a new tab)Website ↗ (opens in a new tab)Repository on GitHub ↗ (opens in a new tab)
335 out of 100

duix-avatar

Offline talking-avatar video generator with voice cloning and lip-sync

16.2k starscustom licenselast commit Apr 2026

Duix.Avatar clones a person's face and voice from a short video, then renders lip-synced videos driven by text or audio. It runs as an Electron client plus Docker services for ASR (fun-asr), TTS (fish-speech-ziming) and video synthesis, all on a local NVIDIA GPU. The services expose local HTTP endpoints for voice synthesis (port 18180) and video synthesis (port 8383).

Strengths

  • Runs fully offline; all compute is local, so no footage leaves the machine
  • Exposes local HTTP APIs for audio synthesis (18180) and video synthesis (8383)
  • Scripts supported in eight languages, including English, Chinese, Japanese and Spanish
  • Clients for Windows (installer) and Ubuntu 22.04 (AppImage); lite compose file available

Weaknesses

  • Requires an NVIDIA GPU; services will not start without one
  • Heavy setup: about 70GB of image downloads and 100GB free disk recommended
  • Non-real-time only; interactive avatars are routed to the paid duix.com service
  • Custom license: enterprises above 100,000 users or 10M USD revenue need a commercial agreement
  • RAM ≥ 32 GB
  • GPU required
  • Compose
  • Needs Docker, NVIDIA Container Toolkit, Node.js 18, guiji2025/fun-asr, guiji2025/fish-speech-ziming, guiji2025/duix.avatar
  • Models: fun-asr, fish-speech-ziming
Documentation ↗ (opens in a new tab)Website ↗ (opens in a new tab)Repository on GitHub ↗ (opens in a new tab)