AWhat changed
Combined coarse self-supervised semantic tokens with fine neural-codec tokens in hierarchical language modeling, generating prompted speech and piano continuations without transcript-based supervision.
Scientific scope
Combined coarse self-supervised semantic tokens with fine neural-codec tokens in hierarchical language modeling, generating prompted speech and piano continuations without transcript-based supervision. No transcript requirement applies to the described audio-generation training, not to every inherited model's complete provenance. Plausible continuations need not be factually correct or authorized speaker reproductions; paper first appeared 2022, cited revision is 2023.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. No transcript requirement applies to the described audio-generation training, not to every inherited model's complete provenance. Plausible continuations need not be factually correct or authorized speaker reproductions; paper first appeared 2022, cited revision is 2023.
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.
- AudioLM: a Language Modeling Approach to Audio Generation
- Borsos et al.: AudioLM semantic and acoustic token generation — author/company project record
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