总结各种创作类型大模型

[imoldpan]

文生对话

• VASA-1 - Microsoft Research

文生图

• Introducing Stable Cascade — Stability AI
• ConsiStory - https://twitter.com/xiaohuggg/status/1758131675094274089

音乐

• MusicGen

视频

• https://boximator.github.io/
• Sora
• https://github.com/Doubiiu/DynamiCrafter

换口型

TTS

• https://github.com/huggingface/parler-tts
• https://github.com/openai/whisper
• https://huggingface.co/spaces/nvidia/canary-1b

参考

• https://huggingface.co/spaces/hf-audio/open_asr_leaderboard