Kimi-Audio, an open-source audio foundation model excelling in audio understanding, generation, and conversation

Official DeiT repository

LPIPS metric. pip install lpips

Minimal keyword extraction with BERT

3D ResNets for Action Recognition (CVPR 2018)

State-of-the-art deep learning based audio codec supporting both mono 24 kHz audio and stereo 48 kHz audio.

🚀 Efficient implementations of state-of-the-art linear attention models

OpenMMLab Pre-training Toolbox and Benchmark

Scenic: A Jax Library for Computer Vision Research and Beyond

Vector (and Scalar) Quantization, in Pytorch

Whisper realtime streaming for long speech-to-text transcription and translation

Towhee is a framework that is dedicated to making neural data processing pipelines simple and fast.

open-source multimodal large language model that can hear, talk while thinking. Featuring real-time end-to-end speech input and streaming audio output conversational capabilities.

[EMNLP 2023 Demo] Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding

Pythonic bindings for FFmpeg's libraries.

View model summaries in PyTorch!

Retinaface get 80.99% in widerface hard val using mobilenet0.25.

[ICLR 2023] Official implementation of the paper "DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection"

GeneFace: Generalized and High-Fidelity 3D Talking Face Synthesis; ICLR 2023; Official code

A Repository for Single- and Multi-modal Speaker Verification, Speaker Recognition and Speaker Diarization

✨✨[NeurIPS 2025] VITA-1.5: Towards GPT-4o Level Real-Time Vision and Speech Interaction

This library provides common speech features for ASR including MFCCs and filterbank energies.

label-smooth, amsoftmax, partial-fc, focal-loss, triplet-loss, lovasz-softmax. Maybe useful

Official PyTorch code for "BAM: Bottleneck Attention Module (BMVC2018)" and "CBAM: Convolutional Block Attention Module (ECCV2018)"

PyTorch implementation of VALL-E(Zero-Shot Text-To-Speech), Reproduced Demo https://lifeiteng.github.io/valle/index.html

The implementation of DeBERTa

Speech Recognition using DeepSpeech2.

An implementation of Microsoft's "FastSpeech 2: Fast and High-Quality End-to-End Text to Speech"

本项目基于SadTalkers实现视频唇形合成的Wav2lip。通过以视频文件方式进行语音驱动生成唇形，设置面部区域可配置的增强方式进行合成唇形（人脸）区域画面增强，提高生成唇形的清晰度。使用DAIN 插帧的DL算法对生成视频进行补帧，补充帧间合成唇形的动作过渡，使合成的唇形更为流畅、真实以及自然。

🔓 Lip Reading - Cross Audio-Visual Recognition using 3D Architectures