Dec 13, 11:05 PM

InternLM-XComposer2.5-OmniLive Launches AI System for Streaming Interactions with Real-time Visual Understanding and Natural Voice Interaction

InternLM-XComposer2.5-OmniLive has been introduced as a comprehensive multimodal system designed for long-term streaming video and audio interactions. This system features real-time visual and auditory understanding, long-term memory formation, and natural voice interaction capabilities. The development of InternLM-XComposer2.5-OmniLive involved researchers from the Shanghai Artificial Intelligence Laboratory, the Chinese University of Hong Kong, Fudan University, and the University of Science and Technology. Additionally, a related model, OmniAudio-2.6B, has been announced as the world's fastest and most efficient audio-language model, capable of processing both text and audio inputs with minimal latency, built upon Google's Gemma-2-2b and Whisper turbo technologies.

#OmniLive #Shanghai Artificial Intelligence Laboratory #Fudan University #OmniAudio #Google #Whisper

Written with ChatGPT (GPT-4o mini).

Sources

Additional media

Image #1 for story internlm-xcomposer2-5-omnilive-launches-ai-system-streaming-interactions-real-9a08f0bd

Image #2 for story internlm-xcomposer2-5-omnilive-launches-ai-system-streaming-interactions-real-9a08f0bd

Image #3 for story internlm-xcomposer2-5-omnilive-launches-ai-system-streaming-interactions-real-9a08f0bd

Image #4 for story internlm-xcomposer2-5-omnilive-launches-ai-system-streaming-interactions-real-9a08f0bd

Image #5 for story internlm-xcomposer2-5-omnilive-launches-ai-system-streaming-interactions-real-9a08f0bd