Search papers, labs, and topics across Lattice.
The paper introduces CineSRD, a multimodal framework for speaker diarization in open-world visual media, leveraging visual anchor clustering for initial speaker registration and an audio language model for speaker turn detection and refinement. This approach addresses challenges like long-form video, numerous speakers, and audiovisual asynchrony. The authors also contribute a new speaker diarization benchmark for visual media in both Chinese and English.
Speaker diarization in movies and TV shows just got a whole lot better, thanks to a new multimodal framework that uses visual cues, speech, and subtitles to handle the chaos of open-world video.
Traditional speaker diarization systems have primarily focused on constrained scenarios such as meetings and interviews, where the number of speakers is limited and acoustic conditions are relatively clean. To explore open-world speaker diarization, we extend this task to the visual media domain, encompassing complex audiovisual programs such as films and TV series. This new setting introduces several challenges, including long-form video understanding, a large number of speakers, cross-modal asynchrony between audio and visual cues, and uncontrolled in-the-wild variability. To address these challenges, we propose Cinematic Speaker Registration&Diarization (CineSRD), a unified multimodal framework that leverages visual, acoustic, and linguistic cues from video, speech, and subtitles for speaker annotation. CineSRD first performs visual anchor clustering to register initial speakers and then integrates an audio language model for speaker turn detection, refining annotations and supplementing unregistered off-screen speakers. Furthermore, we construct and release a dedicated speaker diarization benchmark for visual media that includes Chinese and English programs. Experimental results demonstrate that CineSRD achieves superior performance on the proposed benchmark and competitive results on conventional datasets, validating its robustness and generalizability in open-world visual media settings.
