Enhance Generation Quality of Flow Matching V2A Model via Multi-Step CoT-Like Guidance and Combined Preference Optimization

• URL: http://arxiv.org/abs/2503.22200v1
• Date: Fri, 28 Mar 2025 07:32:14 GMT
• Title: Enhance Generation Quality of Flow Matching V2A Model via Multi-Step CoT-Like Guidance and Combined Preference Optimization
• Authors: Haomin Zhang, Sizhe Shan, Haoyu Wang, Zihao Chen, Xiulong Liu, Chaofan Ding, Xinhan Di,
• Abstract summary: We introduce a multi-stage, multi-modal, end-to-end generative framework with Chain-of-Perform (CoP) guidance learning.<n>We implement a multi-stage training framework that follows step-by-step guidance to ensure the generation of high-quality sound effects.<n>Third, we develop a CoP multi-modal dataset, guided by video, to support step-by-step sound effects generation.
• Score: 11.637343484707014
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Creating high-quality sound effects from videos and text prompts requires precise alignment between visual and audio domains, both semantically and temporally, along with step-by-step guidance for professional audio generation. However, current state-of-the-art video-guided audio generation models often fall short of producing high-quality audio for both general and specialized use cases. To address this challenge, we introduce a multi-stage, multi-modal, end-to-end generative framework with Chain-of-Thought-like (CoT-like) guidance learning, termed Chain-of-Perform (CoP). First, we employ a transformer-based network architecture designed to achieve CoP guidance, enabling the generation of both general and professional audio. Second, we implement a multi-stage training framework that follows step-by-step guidance to ensure the generation of high-quality sound effects. Third, we develop a CoP multi-modal dataset, guided by video, to support step-by-step sound effects generation. Evaluation results highlight the advantages of the proposed multi-stage CoP generative framework compared to the state-of-the-art models on a variety of datasets, with FAD 0.79 to 0.74 (+6.33%), CLIP 16.12 to 17.70 (+9.80%) on VGGSound, SI-SDR 1.98dB to 3.35dB (+69.19%), MOS 2.94 to 3.49(+18.71%) on PianoYT-2h, and SI-SDR 2.22dB to 3.21dB (+44.59%), MOS 3.07 to 3.42 (+11.40%) on Piano-10h.

Related papers

• DeepAudio-V1:Towards Multi-Modal Multi-Stage End-to-End Video to Speech and Audio Generation [6.315946909350621]
  We propose an end-to-end multi-modal generation framework that simultaneously produces speech and audio based on video and text conditions. The proposed framework, DeepAudio, consists of a video-to-audio (V2A) module, a text-to-speech (TTS) module, and a dynamic mixture of modality fusion (MoF) module. In the evaluation, our framework achieves comparable results in the comparison with state-of-the-art models for the video-audio and text-speech benchmarks.
  arXiv  Detail & Related papers  (2025-03-28T09:29:08Z)
• DeepSound-V1: Start to Think Step-by-Step in the Audio Generation from Videos [4.452513686760606]
  We propose a framework for audio generation from videos, leveraging the internal chain-of-thought (CoT) of a multi-modal large language model (MLLM)<n>A corresponding multi-modal reasoning dataset is constructed to facilitate the learning of initial reasoning in audio generation.<n>In experiments, we demonstrate the effectiveness of the proposed framework in reducing misalignment (voice-over) in generated audio.
  arXiv  Detail & Related papers  (2025-03-28T07:56:19Z)
• Baichuan-Omni-1.5 Technical Report [78.49101296394218]
  Baichuan- Omni-1.5 is an omni-modal model that not only has omni-modal understanding capabilities but also provides end-to-end audio generation capabilities.<n>We establish a comprehensive data cleaning and synthesis pipeline for multimodal data, obtaining about 500B high-quality data.<n>Second, an audio-tokenizer has been designed to capture both semantic and acoustic information from audio, enabling seamless integration and enhanced compatibility with MLLM.
  arXiv  Detail & Related papers  (2025-01-26T02:19:03Z)
• Frieren: Efficient Video-to-Audio Generation Network with Rectified Flow Matching [51.70360630470263]
  Video-to-audio (V2A) generation aims to synthesize content-matching audio from silent video.<n>We propose Frieren, a V2A model based on rectified flow matching.<n>Experiments indicate that Frieren achieves state-of-the-art performance in both generation quality and temporal alignment.
  arXiv  Detail & Related papers  (2024-06-01T06:40:22Z)
• Seeing and Hearing: Open-domain Visual-Audio Generation with Diffusion Latent Aligners [69.70590867769408]
  Video and audio content creation serves as the core technique for the movie industry and professional users. Existing diffusion-based methods tackle video and audio generation separately, which hinders the technique transfer from academia to industry. In this work, we aim at filling the gap, with a carefully designed optimization-based framework for cross-visual-audio and joint-visual-audio generation.
  arXiv  Detail & Related papers  (2024-02-27T17:57:04Z)
• Efficient Multiscale Multimodal Bottleneck Transformer for Audio-Video Classification [6.341420717393898]
  We develop a novel multiscale audio Transformer (MAT) and a multiscale video Transformer (MMT) The proposed MAT significantly outperforms AST [28] by 22.2%, 4.4% and 4.7% on three public benchmark datasets. It is about 3% more efficient based on the number of FLOPs and 9.8% more efficient based on GPU memory usage.
  arXiv  Detail & Related papers  (2024-01-08T17:02:25Z)
• MM-Diffusion: Learning Multi-Modal Diffusion Models for Joint Audio and Video Generation [70.74377373885645]
  We propose the first joint audio-video generation framework that brings engaging watching and listening experiences simultaneously. MM-Diffusion consists of a sequential multi-modal U-Net for a joint denoising process by design. Experiments show superior results in unconditional audio-video generation, and zero-shot conditional tasks.
  arXiv  Detail & Related papers  (2022-12-19T14:11:52Z)
• A study on joint modeling and data augmentation of multi-modalities for audio-visual scene classification [64.59834310846516]
  We propose two techniques, namely joint modeling and data augmentation, to improve system performances for audio-visual scene classification (AVSC) Our final system can achieve the best accuracy of 94.2% among all single AVSC systems submitted to DCASE 2021 Task 1b.
  arXiv  Detail & Related papers  (2022-03-07T07:29:55Z)
• VATT: Transformers for Multimodal Self-Supervised Learning from Raw Video, Audio and Text [60.97904439526213]
  Video-Audio-Text Transformer (VATT) takes raw signals as inputs and extracts multimodal representations that are rich enough to benefit a variety of downstream tasks. We train VATT end-to-end from scratch using multimodal contrastive losses and evaluate its performance by the downstream tasks of video action recognition, audio event classification, image classification, and text-to-video retrieval.
  arXiv  Detail & Related papers  (2021-04-22T17:07:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.