Graph-based Label Propagation For Semi-supervised Speaker Identification
2021 · Long Chen, Venkatesh Ravichandran, Andreas Stolcke
Abstract
Speaker identification in the household scenario (e.g., for smart speakers) is typically based on only a few enrollment utterances but a much larger set of unlabeled data, suggesting semisupervised learning to improve speaker profiles. We propose a graph-based semi-supervised learning approach for speaker identification in the household scenario, to leverage the unlabeled speech samples. In contrast to most of the works in speaker recognition that focus on speaker-discriminative embeddings, this work focuses on speaker label inference (scoring). Given a pre-trained embedding extractor, graph-based learning allows us to integrate information about both labeled and unlabeled utterances. Considering each utterance as a graph node, we represent pairwise utterance similarity scores as edge weights. Graphs are constructed per household, and speaker identities are propagated to unlabeled nodes to optimize a global consistency criterion. We show in experiments on the VoxCeleb dataset that this
Authors
(none)
Tags
Stats
Related papers
- Graph-based Multi-view Fusion And Local Adaptation: Mitigating Within-household Confusability For Speaker Identification (2022)2.26
- Improving Speaker Identification For Shared Devices By Adapting Embeddings To Speaker Subsets (2021)4.52
- Learning Speaker Representation With Semi-supervised Learning Approach For Speaker Profiling (2021)0.00
- Improving Text-independent Speaker Verification With Auxiliary Speakers Using Graph (2021)0.00
- Hypergraph Based Semi-supervised Learning Algorithms Applied To Speech Recognition Problem: A Novel Approach (2018)0.00
- Speaker Recognition Using Isomorphic Graph Attention Network Based Pooling On Self-supervised Representation (2023)5.84
- Graph Attention Networks For Speaker Verification (2020)9.23
- Graph Convolutional Network Based Semi-supervised Learning On Multi-speaker Meeting Data (2022)7.50