ASAICVLGSDOct 15, 2024

Mini-Omni2: Towards Open-source GPT-4o with Vision, Speech and Duplex Capabilities

arXiv:2410.11190v399 citationsh-index: 2Has Code
Originality Synthesis-oriented
AI Analysis

This work addresses the problem of replicating advanced multi-modal capabilities for the open-source community, though it is incremental as it builds on existing methods to approximate GPT-4o's functionalities.

The paper tackles the challenge of training a unified multi-modal model like GPT-4o by introducing Mini-Omni2, a visual-audio assistant that provides real-time, end-to-end voice responses to vision and audio queries, achieving performance in individual modalities through a three-stage training process on a limited dataset.

GPT-4o, an all-encompassing model, represents a milestone in the development of large multi-modal language models. It can understand visual, auditory, and textual modalities, directly output audio, and support flexible duplex interaction. Models from the open-source community often achieve some functionalities of GPT-4o, such as visual understanding and voice chat. Nevertheless, training a unified model that incorporates all modalities is challenging due to the complexities of multi-modal data, intricate model architectures, and training processes. In this paper, we introduce Mini-Omni2, a visual-audio assistant capable of providing real-time, end-to-end voice responses to visoin and audio queries. By integrating pretrained visual and auditory encoders, Mini-Omni2 maintains performance in individual modalities. We propose a three-stage training process to align modalities, allowing the language model to handle multi-modal inputs and outputs after training on a limited dataset. For interaction, we introduce a command-based interruption mechanism, enabling more flexible interaction with users. To the best of our knowledge, Mini-Omni2 is one of the closest reproductions of GPT-4o, which have similar form of functionality, and we hope it can offer valuable insights for subsequent research.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes