PersonaTalk is a two-stage framework based on attention mechanisms for achieving high-fidelity and personalized visual dubbing. The technology utilizes a style-sensitive audio encoding module and a dual-attention facial renderer, enabling accurate lip-syncing while maintaining and highlighting the speaker's 'personality.' It captures the unique speaking style of the speaker and preserves facial details, which poses a significant challenge for audio-driven visual dubbing. The main advantages of PersonaTalk include high visual quality, precise lip-syncing, and personality retention, functioning as a universal framework capable of matching the performance of methods tailored to specific characters.