AlignDiT: Multimodal Aligned Diffusion Transformer for Synchronized Speech Generation

• URL: http://arxiv.org/abs/2504.20629v1
• Date: Tue, 29 Apr 2025 10:56:24 GMT
• Title: AlignDiT: Multimodal Aligned Diffusion Transformer for Synchronized Speech Generation
• Authors: Jeongsoo Choi, Ji-Hoon Kim, Kim Sung-Bin, Tae-Hyun Oh, Joon Son Chung,
• Abstract summary: multimodal-to-speech task has gained increasing attention due to its wide range of applications, such as film production, dubbing, and virtual avatars.<n>Existing methods still suffer from limitations in speech intelligibility, audio-video synchronization, speech naturalness, and voice similarity to the reference speaker.<n>We propose AlignDiT, a multimodal Aligned Diffusion Transformer that generates accurate, synchronized, and natural-sounding speech from aligned multimodal inputs.
• Score: 41.74261260212531
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In this paper, we address the task of multimodal-to-speech generation, which aims to synthesize high-quality speech from multiple input modalities: text, video, and reference audio. This task has gained increasing attention due to its wide range of applications, such as film production, dubbing, and virtual avatars. Despite recent progress, existing methods still suffer from limitations in speech intelligibility, audio-video synchronization, speech naturalness, and voice similarity to the reference speaker. To address these challenges, we propose AlignDiT, a multimodal Aligned Diffusion Transformer that generates accurate, synchronized, and natural-sounding speech from aligned multimodal inputs. Built upon the in-context learning capability of the DiT architecture, AlignDiT explores three effective strategies to align multimodal representations. Furthermore, we introduce a novel multimodal classifier-free guidance mechanism that allows the model to adaptively balance information from each modality during speech synthesis. Extensive experiments demonstrate that AlignDiT significantly outperforms existing methods across multiple benchmarks in terms of quality, synchronization, and speaker similarity. Moreover, AlignDiT exhibits strong generalization capability across various multimodal tasks, such as video-to-speech synthesis and visual forced alignment, consistently achieving state-of-the-art performance. The demo page is available at https://mm.kaist.ac.kr/projects/AlignDiT .

Related papers

• From Faces to Voices: Learning Hierarchical Representations for High-quality Video-to-Speech [26.67378997911053]
  The objective of this study is to generate high-quality speech from silent talking face videos.<n>We propose a novel video-to-speech system that bridges the modality gap between silent video and multi-faceted speech.<n>Our method achieves exceptional generation quality comparable to real utterances.
  arXiv  Detail & Related papers  (2025-03-21T09:02:38Z)
• HOP: Heterogeneous Topology-based Multimodal Entanglement for Co-Speech Gesture Generation [42.30003982604611]
  Co-speech gestures are crucial non-verbal cues that enhance speech clarity and strides in human communication.<n>We propose a novel method named HOP for co-speech gesture generation, capturing heterogeneous entanglement between gesture motion, audio rhythm, and text semantics.<n>HOP achieves state-of-the-art offering more natural and expressive co-speech gesture generation.
  arXiv  Detail & Related papers  (2025-03-03T04:47:39Z)
• Unispeaker: A Unified Approach for Multimodality-driven Speaker Generation [66.49076386263509]
  This paper introduces UniSpeaker, a unified approach for multimodality-driven speaker generation. We propose a unified voice aggregator based on KV-Former, applying soft contrastive loss to map diverse voice description modalities into a shared voice space. UniSpeaker is evaluated across five tasks using the MVC benchmark, and the experimental results demonstrate that UniSpeaker outperforms previous modality-specific models.
  arXiv  Detail & Related papers  (2025-01-11T00:47:29Z)
• DiTCtrl: Exploring Attention Control in Multi-Modal Diffusion Transformer for Tuning-Free Multi-Prompt Longer Video Generation [54.30327187663316]
  DiTCtrl is a training-free multi-prompt video generation method under MM-DiT architectures for the first time.<n>We analyze MM-DiT's attention mechanism, finding that the 3D full attention behaves similarly to that of the cross/self-attention blocks in the UNet-like diffusion models.<n>Based on our careful design, the video generated by DiTCtrl achieves smooth transitions and consistent object motion given multiple sequential prompts.
  arXiv  Detail & Related papers  (2024-12-24T18:51:19Z)
• Visatronic: A Multimodal Decoder-Only Model for Speech Synthesis [13.702423348269155]
  We propose a new task -- generating speech from videos of people and their transcripts (VTTS) -- to motivate new techniques for multimodal speech generation. We present a decoder-only multimodal model for this task, which we call Visatronic. It embeds vision, text and speech directly into the common subspace of a transformer model and uses an autoregressive loss to learn a generative model of discretized mel-spectrograms conditioned on speaker videos and transcripts of their speech.
  arXiv  Detail & Related papers  (2024-11-26T18:57:29Z)
• Towards Multi-Task Multi-Modal Models: A Video Generative Perspective [5.495245220300184]
  This thesis chronicles our endeavor to build multi-task models for generating videos and other modalities under diverse conditions. We unveil a novel approach to mapping bidirectionally between visual observation and interpretable lexical terms. Our scalable visual token representation proves beneficial across generation, compression, and understanding tasks.
  arXiv  Detail & Related papers  (2024-05-26T23:56:45Z)
• VioLA: Unified Codec Language Models for Speech Recognition, Synthesis, and Translation [91.39949385661379]
  VioLA is a single auto-regressive Transformer decoder-only network that unifies various cross-modal tasks involving speech and text. We first convert all the speech utterances to discrete tokens using an offline neural encoder. We further integrate task IDs (TID) and language IDs (LID) into the proposed model to enhance the modeling capability of handling different languages and tasks.
  arXiv  Detail & Related papers  (2023-05-25T14:39:47Z)
• Learning to Listen: Modeling Non-Deterministic Dyadic Facial Motion [89.01668641930206]
  We present a framework for modeling interactional communication in dyadic conversations. We autoregressively output multiple possibilities of corresponding listener motion. Our method organically captures the multimodal and non-deterministic nature of nonverbal dyadic interactions.
  arXiv  Detail & Related papers  (2022-04-18T17:58:04Z)
• Multi Modal Adaptive Normalization for Audio to Video Generation [18.812696623555855]
  We propose a multi-modal adaptive normalization(MAN) based architecture to synthesize a talking person video of arbitrary length using as input: an audio signal and a single image of a person. The architecture uses the multi-modal adaptive normalization, keypoint heatmap predictor, optical flow predictor and class activation map[58] based layers to learn movements of expressive facial components.
  arXiv  Detail & Related papers  (2020-12-14T07:39:45Z)
• MultiSpeech: Multi-Speaker Text to Speech with Transformer [145.56725956639232]
  Transformer-based text to speech (TTS) model (e.g., Transformer TTSciteli 2019neural, FastSpeechciteren 2019fastspeech) has shown the advantages of training and inference efficiency over RNN-based model. We develop a robust and high-quality multi-speaker Transformer TTS system called MultiSpeech, with several specially designed components/techniques to improve text-to-speech alignment.
  arXiv  Detail & Related papers  (2020-06-08T15:05:28Z)

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.