Self-supervised Fine-tuning for Improved Content Representations by Speaker-invariant Clustering.
Heng-Jui ChangAlexander H. LiuJames R. GlassPublished in: CoRR (2023)
Keyphrases
- fine tuning
- fine tuned
- viable alternative
- clustering algorithm
- k means
- clustering method
- automatically discovering
- data clustering
- invariant representations
- fine tune
- information theoretic
- multiple representations
- cluster analysis
- feature selection
- hierarchical clustering
- categorical data
- multimedia
- audio visual
- spectral clustering
- data sets
- affine transformation
- self organizing maps
- web documents
- speaker verification
- data representations
- higher level
- web pages
- document clustering
- search engine