{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:Q33I4SSQC4RQ3WWYWW46UMJZZ7","short_pith_number":"pith:Q33I4SSQ","schema_version":"1.0","canonical_sha256":"86f68e4a5017230ddad8b5b9ea3139cff5b329528a72deec9141ea7af93631e1","source":{"kind":"arxiv","id":"2305.11072","version":1},"attestation_state":"computed","paper":{"title":"Self-supervised Fine-tuning for Improved Content Representations by Speaker-invariant Clustering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Alexander H. Liu, Heng-Jui Chang, James Glass","submitted_at":"2023-05-18T15:59:36Z","abstract_excerpt":"Self-supervised speech representation models have succeeded in various tasks, but improving them for content-related problems using unlabeled data is challenging. We propose speaker-invariant clustering (Spin), a novel self-supervised learning method that clusters speech representations and performs swapped prediction between the original and speaker-perturbed utterances. Spin disentangles speaker information and preserves content representations with just 45 minutes of fine-tuning on a single GPU. Spin improves pre-trained networks and outperforms prior methods in speech recognition and acous"},"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":"2305.11072","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-18T15:59:36Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"2b242e3ca8e2b22290a2975a7cf915028dce9407e3e3c209e4dbd76b5f59b7ec","abstract_canon_sha256":"0569412b73d943863a07dda0ec1ed09af32c16635f90fe75b5f7dc4f0e4500e1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:11:30.100055Z","signature_b64":"WnN5ROujndEm+czVX7EaZmQ260V5ASryzr419t5Q037ZtYNWANgZ2zhPVERSE/nKOISYhsQ4/KeZc0ZUd4mUDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"86f68e4a5017230ddad8b5b9ea3139cff5b329528a72deec9141ea7af93631e1","last_reissued_at":"2026-07-05T06:11:30.099662Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:11:30.099662Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-supervised Fine-tuning for Improved Content Representations by Speaker-invariant Clustering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Alexander H. Liu, Heng-Jui Chang, James Glass","submitted_at":"2023-05-18T15:59:36Z","abstract_excerpt":"Self-supervised speech representation models have succeeded in various tasks, but improving them for content-related problems using unlabeled data is challenging. We propose speaker-invariant clustering (Spin), a novel self-supervised learning method that clusters speech representations and performs swapped prediction between the original and speaker-perturbed utterances. Spin disentangles speaker information and preserves content representations with just 45 minutes of fine-tuning on a single GPU. Spin improves pre-trained networks and outperforms prior methods in speech recognition and acous"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.11072","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/2305.11072/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":"2305.11072","created_at":"2026-07-05T06:11:30.099715+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.11072v1","created_at":"2026-07-05T06:11:30.099715+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.11072","created_at":"2026-07-05T06:11:30.099715+00:00"},{"alias_kind":"pith_short_12","alias_value":"Q33I4SSQC4RQ","created_at":"2026-07-05T06:11:30.099715+00:00"},{"alias_kind":"pith_short_16","alias_value":"Q33I4SSQC4RQ3WWY","created_at":"2026-07-05T06:11:30.099715+00:00"},{"alias_kind":"pith_short_8","alias_value":"Q33I4SSQ","created_at":"2026-07-05T06:11:30.099715+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.05586","citing_title":"FreeSVC: Towards Zero-shot Multilingual Singing Voice Conversion","ref_index":18,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Q33I4SSQC4RQ3WWYWW46UMJZZ7","json":"https://pith.science/pith/Q33I4SSQC4RQ3WWYWW46UMJZZ7.json","graph_json":"https://pith.science/api/pith-number/Q33I4SSQC4RQ3WWYWW46UMJZZ7/graph.json","events_json":"https://pith.science/api/pith-number/Q33I4SSQC4RQ3WWYWW46UMJZZ7/events.json","paper":"https://pith.science/paper/Q33I4SSQ"},"agent_actions":{"view_html":"https://pith.science/pith/Q33I4SSQC4RQ3WWYWW46UMJZZ7","download_json":"https://pith.science/pith/Q33I4SSQC4RQ3WWYWW46UMJZZ7.json","view_paper":"https://pith.science/paper/Q33I4SSQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.11072&json=true","fetch_graph":"https://pith.science/api/pith-number/Q33I4SSQC4RQ3WWYWW46UMJZZ7/graph.json","fetch_events":"https://pith.science/api/pith-number/Q33I4SSQC4RQ3WWYWW46UMJZZ7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Q33I4SSQC4RQ3WWYWW46UMJZZ7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Q33I4SSQC4RQ3WWYWW46UMJZZ7/action/storage_attestation","attest_author":"https://pith.science/pith/Q33I4SSQC4RQ3WWYWW46UMJZZ7/action/author_attestation","sign_citation":"https://pith.science/pith/Q33I4SSQC4RQ3WWYWW46UMJZZ7/action/citation_signature","submit_replication":"https://pith.science/pith/Q33I4SSQC4RQ3WWYWW46UMJZZ7/action/replication_record"}},"created_at":"2026-07-05T06:11:30.099715+00:00","updated_at":"2026-07-05T06:11:30.099715+00:00"}