VieSpeaker: A Large-Scale Vietnamese Speaker Recognition Dataset Beyond Visual Dependency
This work addresses the lack of large-scale, acoustically diverse Vietnamese speaker recognition data by providing a new dataset and a face-independent construction method.
The authors constructed VieSpeaker, a 902-hour Vietnamese speaker recognition dataset with 4,715 speakers using a face-independent pipeline, and showed that models trained on it achieve improved robustness and generalization over existing Vietnamese datasets.
Speaker recognition has advanced rapidly with large-scale training datasets, yet Vietnamese remains under-resourced, with existing corpora limited in scale and acoustic diversity. Most large-scale datasets rely on facial cues to link speech with speaker identities, restricting data collection to recordings where speakers appear on camera. We propose a face-independent dataset construction pipeline and introduce VieSpeaker, a large-scale Vietnamese speaker recognition dataset. Our approach leverages textual metadata and large language model reasoning to infer speaker identities from transcripts and contextual information. VieSpeaker contains approximately 902 hours of speech from 4,715 speakers. Experiments show that models trained on VieSpeaker achieve improved robustness and generalization compared to existing Vietnamese datasets. This work demonstrates the feasibility of face-independent dataset construction and provides a new direction for building large-scale speech resources.