Blog Tags GitHub

audio-generation

6 posts tagged with "audio-generation"

Movie Gen: 媒体基础模型全家桶

Meta 的 Movie Gen 媒体基础模型全家桶：30B 文生视频 + 13B 视频音频生成，Flow Matching + LLaMa3 风格 Transformer，支持 1080p HD 视频、同步音频、个性化与指令式视频编辑，多任务达到 SOTA。

diffusion-model video-generation audio-generation large-model-training flow-matching

Qwen3-Omni Technical Report

阿里巴巴Qwen团队的统一多模态模型，在文本、图像、音频、视频上无性能退化

audio mixture-of-experts diffusion-transformer llm-training audio-generation

Fish Audio S2 Technical Report

开源可控文本到语音系统，支持多说话人、多轮对话和自然语言指令控制

audio reinforcement-learning vision-language audio-generation

DiffServe: Efficiently Serving Text-to-Image Diffusion Models with Query-Aware Model Scaling

基于查询感知模型扩展的文本到图像扩散模型高效服务系统

diffusion-model inference-serving image-generation benchmark audio-generation

FastTree: Optimizing Attention Kernel and Runtime for Tree-Structured LLM Inference

优化树结构LLM推理的注意力内核和运行时

efficient-attention llm-inference kv-cache audio-generation

SkyReels-V4: Multi-modal Video-Audio Generation, Inpainting and Editing model

首个同时支持多模态输入、联合视频-音频生成、统一生成/修复/编辑的视频基础模型，采用双流 MMDiT 架构，支持 1080p 32FPS 15秒电影级视频生成

audio diffusion-model sparse-attention video-generation vlm mllm audio-generation