StyleDubber: Towards Multi-Scale Style Learning for Movie Dubbing

TL;DR

StyleDubber introduces a phoneme-level approach to movie dubbing that improves temporal alignment, pronunciation accuracy, and emotional consistency by leveraging multi-scale style learning and lip synchronization.

Contribution

It proposes a novel phoneme-level style learning framework for movie dubbing, enhancing lip sync and emotional expression over existing frame-level models.

Findings

Outperforms state-of-the-art on V2C and Grid benchmarks.

Improves lip sync accuracy and emotional consistency.

Enhances pronunciation stability and style transfer quality.

Abstract

Given a script, the challenge in Movie Dubbing (Visual Voice Cloning, V2C) is to generate speech that aligns well with the video in both time and emotion, based on the tone of a reference audio track. Existing state-of-the-art V2C models break the phonemes in the script according to the divisions between video frames, which solves the temporal alignment problem but leads to incomplete phoneme pronunciation and poor identity stability. To address this problem, we propose StyleDubber, which switches dubbing learning from the frame level to phoneme level. It contains three main components: (1) A multimodal style adaptor operating at the phoneme level to learn pronunciation style from the reference audio, and generate intermediate representations informed by the facial emotion presented in the video; (2) An utterance-level style learning module, which guides both the mel-spectrogram decoding and the refining processes from the intermediate embeddings to improve the overall style expression; And (3) a phoneme-guided lip aligner to maintain lip sync. Extensive experiments on two of the primary benchmarks, V2C and Grid, demonstrate the favorable performance of the proposed method as compared to the current stateof-the-art. The code will be made available at https://github.com/GalaxyCong/StyleDubber.