最有价值项目

YesBut - Multimodal Satire Comprehension Dataset

模式识别课设代码：图文生成（CLIP+DALLE+BriVL）

A multimodal database engine written in Nim

小红书图文笔记 Agent Skill，支持看图理解、主动提问、智能选题、实时热梗搜索、标题正文、标签评论钩子和风险检查。

Local LLM Discord Bot

Code for "MIFNet: Learning Modality-Invariant Features for Generalizable Multimodal Image Matching", TIP2025

The code for "VISTA: Enhancing Long-Duration and High-Resolution Video Understanding by VIdeo SpatioTemporal Augmentation" [CVPR2025]

Implementation of "With a Little Help from my Temporal Context: Multimodal Egocentric Action Recognition, BMVC, 2021" in PyTorch

ABC: Achieving Better Control of Multimodal Embeddings using VLMs [TMLR2025]

IA na Prática: LLM, RAG, MCP, Agents, Function Calling, Multimodal, TTS/STT e mais

Run Ornith, Qwen, Nemotron & DeepSeek optimized on AMD Strix Halo — unified OpenAI-compatible server with tuned ROCmFP4 quants, validated 262K context, vision support & hot-swap model zoo (iGPU + NPU)

React component and hook for speech recognition, with voice commands and pluggable STT engines

Automatic Speech Recognition tool

MCP server for transcript processing — formatting, contextual repair & smart summarization with deep-thinking LLMs

HTK Toolkit with Linux 64 bit and Docker support

An Android app using whisper.cpp for voice-to-text transcription

A Flutter-based space exploration app with NASA APIs, ISS tracker, space missions, and AI-powered chatbot.

🎙️ Translate as you speak using Google Chrome's Web Speech API for speech recognition.

Magic Media is an award-winning experimental native iOS/iPadOS application based on Augmented Reality (AR) that focuses on augmenting physical forms of media such as magazines, dollar bills, instruction manuals, etc. It delivers its users various options to interact with several sections of the physical media with images/text and then convert them into videos or 3D models specifically. This will help the user understand the information better in various ways of interaction, which is majorly made possible by AR. It also supports voice commands to help the user navigate through the app and have a more friendly and immersive experience.

An implementation for "Conformer: Convolution-augmented Transformer for Speech Recognition" Paper