{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JAOKW7PIDJZ6GYFJHWXIPV647O","short_pith_number":"pith:JAOKW7PI","schema_version":"1.0","canonical_sha256":"481cab7de81a73e360a93dae87d7dcfbbb8bd29b2763df8a584fe7186a81d321","source":{"kind":"arxiv","id":"2508.05207","version":1},"attestation_state":"computed","paper":{"title":"SpectroStream: A Versatile Neural Codec for General Audio","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Brian McWilliams, Kehang Han, Marco Tagliasacchi, Yunpeng Li, Zalan Borsos","submitted_at":"2025-08-07T09:44:00Z","abstract_excerpt":"We propose SpectroStream, a full-band multi-channel neural audio codec. Successor to the well-established SoundStream, SpectroStream extends its capability beyond 24 kHz monophonic audio and enables high-quality reconstruction of 48 kHz stereo music at bit rates of 4--16 kbps. This is accomplished with a new neural architecture that leverages audio representation in the time-frequency domain, which leads to better audio quality especially at higher sample rate. The model also uses a delayed-fusion strategy to handle multi-channel audio, which is crucial in balancing per-channel acoustic qualit"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2508.05207","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2025-08-07T09:44:00Z","cross_cats_sorted":["cs.AI","eess.AS"],"title_canon_sha256":"aef45fbcadef481ebb093f3b8c59611c168625d14228b931fe263eda75a7736b","abstract_canon_sha256":"29cffa8e13508cd64b2250e5e77d8241f6da1cd5a20c6d9eba272de7d85fd8ec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:50:08.902014Z","signature_b64":"cSlMFy/XyU3XYs71N/odxHlGnQ82veope/Xg4oK6MtiqzAN0bdScTK5/UsXO/xGNdg56gRWnoaN/oTYsjnXoCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"481cab7de81a73e360a93dae87d7dcfbbb8bd29b2763df8a584fe7186a81d321","last_reissued_at":"2026-07-05T11:50:08.901557Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:50:08.901557Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SpectroStream: A Versatile Neural Codec for General Audio","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Brian McWilliams, Kehang Han, Marco Tagliasacchi, Yunpeng Li, Zalan Borsos","submitted_at":"2025-08-07T09:44:00Z","abstract_excerpt":"We propose SpectroStream, a full-band multi-channel neural audio codec. Successor to the well-established SoundStream, SpectroStream extends its capability beyond 24 kHz monophonic audio and enables high-quality reconstruction of 48 kHz stereo music at bit rates of 4--16 kbps. This is accomplished with a new neural architecture that leverages audio representation in the time-frequency domain, which leads to better audio quality especially at higher sample rate. The model also uses a delayed-fusion strategy to handle multi-channel audio, which is crucial in balancing per-channel acoustic qualit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.05207","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2508.05207/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2508.05207","created_at":"2026-07-05T11:50:08.901613+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.05207v1","created_at":"2026-07-05T11:50:08.901613+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.05207","created_at":"2026-07-05T11:50:08.901613+00:00"},{"alias_kind":"pith_short_12","alias_value":"JAOKW7PIDJZ6","created_at":"2026-07-05T11:50:08.901613+00:00"},{"alias_kind":"pith_short_16","alias_value":"JAOKW7PIDJZ6GYFJ","created_at":"2026-07-05T11:50:08.901613+00:00"},{"alias_kind":"pith_short_8","alias_value":"JAOKW7PI","created_at":"2026-07-05T11:50:08.901613+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27320","citing_title":"Elastic Time: Dynamic Frame Rate Bottlenecks for Neural Audio Coding","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06357","citing_title":"F3-Tokenizer: Taming Audio Autoencoder Latents for Understanding and Generation","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17085","citing_title":"Taming Audio VAEs via Target-KL Regularization","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JAOKW7PIDJZ6GYFJHWXIPV647O","json":"https://pith.science/pith/JAOKW7PIDJZ6GYFJHWXIPV647O.json","graph_json":"https://pith.science/api/pith-number/JAOKW7PIDJZ6GYFJHWXIPV647O/graph.json","events_json":"https://pith.science/api/pith-number/JAOKW7PIDJZ6GYFJHWXIPV647O/events.json","paper":"https://pith.science/paper/JAOKW7PI"},"agent_actions":{"view_html":"https://pith.science/pith/JAOKW7PIDJZ6GYFJHWXIPV647O","download_json":"https://pith.science/pith/JAOKW7PIDJZ6GYFJHWXIPV647O.json","view_paper":"https://pith.science/paper/JAOKW7PI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.05207&json=true","fetch_graph":"https://pith.science/api/pith-number/JAOKW7PIDJZ6GYFJHWXIPV647O/graph.json","fetch_events":"https://pith.science/api/pith-number/JAOKW7PIDJZ6GYFJHWXIPV647O/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JAOKW7PIDJZ6GYFJHWXIPV647O/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JAOKW7PIDJZ6GYFJHWXIPV647O/action/storage_attestation","attest_author":"https://pith.science/pith/JAOKW7PIDJZ6GYFJHWXIPV647O/action/author_attestation","sign_citation":"https://pith.science/pith/JAOKW7PIDJZ6GYFJHWXIPV647O/action/citation_signature","submit_replication":"https://pith.science/pith/JAOKW7PIDJZ6GYFJHWXIPV647O/action/replication_record"}},"created_at":"2026-07-05T11:50:08.901613+00:00","updated_at":"2026-07-05T11:50:08.901613+00:00"}