talking-face

[CVPR 2023] SadTalker：Learning Realistic 3D Motion Coefficients for Stylized Audio-Driven Single Image Talking Face Animation

[CVPR 2024] This is the official source for our paper "SyncTalk: The Devil is in the Synchronization for Talking Head Synthesis"

Wav2Lip UHQ extension for Automatic1111

Wav2Lip version 288 and pipeline to train

code for paper "Audio2Head: Audio-driven One-shot Talking-head Generation with Natural Head Motion" in the conference of IJCAI 2021

Parallel and High-Fidelity Text-to-Lip Generation; AAAI 2022 ; Official code

DoyenTalker uses deep learning techniques to generate personalized avatar videos that speak user-provided text in a specified voice. The system utilizes Coqui TTS for text-to-speech generation, along with various face rendering and animation techniques to create a video where the given avatar articulates the speech.

Database of "Learning to Predict Salient Faces: A Novel Visual-Audio Saliency Model", ECCV 2020

Thin plate spline motion model TPSMM converted to ONNX

[CVPR 2026] IP-Adapter Is All You Need: Towards Fine-Tuning-Free Diffusion-Based Talking Face Generation

SyncNet's modern implementation (Python 3.9~3.13)

CPU-only neural talking-head synthesis. LivePortrait ONNX pipeline (5 stages) with MediaPipe face detection. Apache 2.0. No GPU required.

🎓 Update Talking-Face Research Papers Daily, Now Integrated with LLM Analysis.

Style-Talking is a zero-shot audio-driven digital human system for cloning speaking style from a reference video.

[SIGGRAPH Asia 2022] VideoReTalking: Audio-based Lip Synchronization for Talking Head Video Editing In the Wild

FastAPI-based Text-to-Speech API with Llama integration