Hsu et al.: HuBERT masked prediction of clustered speech units

What changed

Used offline clustering to create hidden-unit targets for masked-region prediction, iterating the target construction to learn acoustic and linguistic representations from unlabeled speech.

Scientific scope

Used offline clustering to create hidden-unit targets for masked-region prediction, iterating the target construction to learn acoustic and linguistic representations from unlabeled speech. Distinct from wav2vec2's contrastive/quantization setup; masked language modeling and clustering are inherited ideas. Benchmark improvements are source-reported, not guarantees across languages or downstream tasks.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Distinct from wav2vec2's contrastive/quantization setup; masked language modeling and clustering are inherited ideas. Benchmark improvements are source-reported, not guarantees across languages or downstream tasks.

Contributors

• Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

• HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units
• Hsu et al.: HuBERT masked prediction of clustered speech units — institutional/author source

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction