wav2lip - 技术专题

fay是一个帮助数字人（2.5d、3d、移动、pc、网页）或大语言模型（openai兼容、deepseek）连通业务系统的agent框架。

Real time interactive streaming digital human

PaddlePaddle GAN library, including lots of interesting applications like First-Order motion transfer, Wav2Lip, picture repair, image editing, photo2cartoon, image style transfer, GPEN, and so on.

AIGCPanel 是一个简单易用的一站式AI数字人系统，支持视频合成、声音合成、声音克隆，简化本地模型管理、一键导入和使用AI模型。

Wav2Lip UHQ extension for Automatic1111

High quality Lip sync

PyTorch Implementation for Paper "Emotionally Enhanced Talking Face Generation" (ICCVW'23 and ACM-MMW'23)

lipsync is a simple and updated Python library for lip synchronization, based on Wav2Lip. It synchronizes lips in videos and images based on provided audio, supports CPU/CUDA, and uses caching for faster processing.

Alternative to Flawless AI's TrueSync. Make lips in video match provided audio using the power of Wav2Lip and GFPGAN.

(Windows/Linux/MacOS) Local WebUI with neural network models (Text, Image, Video, 3D, Audio) on python (Gradio interface). Translated on 3 languages

Orchestrating AI for stunning lip-synced videos. Effortless workflow, exceptional results, all in one place.

Local-first, open-source YouTube dubbing: clones the original voice into another language, time-synced. Chatterbox voice cloning + faster-whisper + NLLB, with multi-voice, burned subtitles and optional Wav2Lip lip-sync. Runs free on your own machine via a simple CLI.

Wav2Lip UHQ Improvement with ControlNet 1.1

AI Lip Syncing application, deployed on Streamlit

AI Talking Head: create video from plain text or audio file in minutes, support up to 100+ languages and 350+ voice models.

End-to-end speech-to-speech translation pipeline with voice cloning (RVC) and automatic lip-sync (Wav2Lip).

Create deepfake video by just uploading the original video and specifying the text the character will read

Turn a single theme into a cinematic short film, a local-first pipeline that drafts the script, storyboards each scene, renders the video, and lip-syncs the voiceover, with graceful fallback to cloud providers when needed.

通過少量語音與影片樣本生成高質量的語音與影片克隆 ( AI 人像口白生成 )，並提供多種音頻處理技術來提升音質和真實感。

End-to-end AI digital-human live-streaming platform: create avatars (image / Edge-TTS voice / persona) → LivePortrait base video → Edge-TTS + Wav2Lip lip-sync → DeepSeek LLM live replies → SRS streaming + Next.js viewer. Built with Go (Gin), MariaDB, Redis, RustFS, Docker, and a Python (uv) AI worker (macOS MPS / Linux CUDA / AMD ROCm).