[CVPR 2026] Official implementation of MultiAnimate: Pose-Guided Image Animation Made Extensible

Long-form streaming TTS system for multi-speaker dialogue generation

The python library for real-time communication

ARC-AGI Toolkit

Hallo: Hierarchical Audio-Driven Visual Synthesis for Portrait Image Animation

Official, AWS-supported MCP servers, skills, and plugins to help AI agents build on AWS

A framework for few-shot evaluation of language models.

τ-Bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains

The Full-Duplex Interaction Track of the ICASSP 2026 Human-like Spoken Dialogue Systems Challenge aims to advance the evaluation of full-duplex dialogue systems by in- troducing a dual-channel dial…

A lightweight library for evaluating vision language models.

The repository for Springer IJCV 2025 (LR-ASD: Lightweight and Robust Network for Active Speaker Detection)

A suite of image and video neural tokenizers

[NeurIPS 2025] Let Them Talk: Audio-Driven Multi-Person Conversational Video Generation

Run agents like Hermes, LangChain Deep Agents, and OpenClaw more securely inside NVIDIA OpenShell with managed inference

A fast and soft pattern search for trillion-scale corpora.

Versatile Evaluation of Speech and Audio

Kyutai with an "eye"

Transformer with Local Modeling by Convolution for Speech Separation and Enhancement

MTalk-Bench: Evaluating Speech-to-Speech Models in Multi-Turn Dialogues via Arena-style and Rubrics Protocols

Repository for the SIGDIAL 2026 website

Qwen3-omni is a natively end-to-end, omni-modal LLM developed by the Qwen team at Alibaba Cloud, capable of understanding text, audio, images, and video, as well as generating speech in real time.

Benchmarking scripts for Gaia

Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec.

Code for the paper Hybrid Spectrogram and Waveform Source Separation

Official Implementation of NAACL 2025 Paper: Behavior-SD: Behaviorally Aware Spoken Dialogue Generation with Large Language Models