{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:B45EYZM2KKJGX5O4DYXDL4MITP","short_pith_number":"pith:B45EYZM2","schema_version":"1.0","canonical_sha256":"0f3a4c659a52926bf5dc1e2e35f1889bf871e9cb38977560bbd20ffbdbef4cee","source":{"kind":"arxiv","id":"2110.05752","version":1},"attestation_state":"computed","paper":{"title":"UniSpeech-SAT: Universal Speech Representation Learning with Speaker Aware Pre-Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Chengyi Wang, Furu Wei, Jian Wu, Jinyu Li, Sanyuan Chen, Shujie Liu, Xiangzhan Yu, Yao Qian, Yu Wu, Zhengyang Chen, Zhuo Chen","submitted_at":"2021-10-12T05:43:30Z","abstract_excerpt":"Self-supervised learning (SSL) is a long-standing goal for speech processing, since it utilizes large-scale unlabeled data and avoids extensive human labeling. Recent years witness great successes in applying self-supervised learning in speech recognition, while limited exploration was attempted in applying SSL for modeling speaker characteristics. In this paper, we aim to improve the existing SSL framework for speaker representation learning. Two methods are introduced for enhancing the unsupervised speaker information extraction. First, we apply the multi-task learning to the current SSL fra"},"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":"2110.05752","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-10-12T05:43:30Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"8bf8dda9ce2e5a33d1f0d0dc0ccea98af74b5af604fdb23ec489950ca9ff237f","abstract_canon_sha256":"201fbd7eaad8c89709c3fca8fa58ce5a1059009a64bcd672828484e76bb003bb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:22:06.099787Z","signature_b64":"yFicIm5xCRSxMcf1kuRIg39mtd4ueeQ+4tCJmVYc8oi8Hdb3b7xLtt3R78pkbqEja7D1LwM7d7mX/UwhzLgDBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f3a4c659a52926bf5dc1e2e35f1889bf871e9cb38977560bbd20ffbdbef4cee","last_reissued_at":"2026-07-05T03:22:06.099361Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:22:06.099361Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UniSpeech-SAT: Universal Speech Representation Learning with Speaker Aware Pre-Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Chengyi Wang, Furu Wei, Jian Wu, Jinyu Li, Sanyuan Chen, Shujie Liu, Xiangzhan Yu, Yao Qian, Yu Wu, Zhengyang Chen, Zhuo Chen","submitted_at":"2021-10-12T05:43:30Z","abstract_excerpt":"Self-supervised learning (SSL) is a long-standing goal for speech processing, since it utilizes large-scale unlabeled data and avoids extensive human labeling. Recent years witness great successes in applying self-supervised learning in speech recognition, while limited exploration was attempted in applying SSL for modeling speaker characteristics. In this paper, we aim to improve the existing SSL framework for speaker representation learning. Two methods are introduced for enhancing the unsupervised speaker information extraction. First, we apply the multi-task learning to the current SSL fra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.05752","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/2110.05752/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":"2110.05752","created_at":"2026-07-05T03:22:06.099420+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.05752v1","created_at":"2026-07-05T03:22:06.099420+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.05752","created_at":"2026-07-05T03:22:06.099420+00:00"},{"alias_kind":"pith_short_12","alias_value":"B45EYZM2KKJG","created_at":"2026-07-05T03:22:06.099420+00:00"},{"alias_kind":"pith_short_16","alias_value":"B45EYZM2KKJGX5O4","created_at":"2026-07-05T03:22:06.099420+00:00"},{"alias_kind":"pith_short_8","alias_value":"B45EYZM2","created_at":"2026-07-05T03:22:06.099420+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21052","citing_title":"Backdoor Attacks on Speech Emotion Recognition via TTS-Generated Poisoning","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B45EYZM2KKJGX5O4DYXDL4MITP","json":"https://pith.science/pith/B45EYZM2KKJGX5O4DYXDL4MITP.json","graph_json":"https://pith.science/api/pith-number/B45EYZM2KKJGX5O4DYXDL4MITP/graph.json","events_json":"https://pith.science/api/pith-number/B45EYZM2KKJGX5O4DYXDL4MITP/events.json","paper":"https://pith.science/paper/B45EYZM2"},"agent_actions":{"view_html":"https://pith.science/pith/B45EYZM2KKJGX5O4DYXDL4MITP","download_json":"https://pith.science/pith/B45EYZM2KKJGX5O4DYXDL4MITP.json","view_paper":"https://pith.science/paper/B45EYZM2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.05752&json=true","fetch_graph":"https://pith.science/api/pith-number/B45EYZM2KKJGX5O4DYXDL4MITP/graph.json","fetch_events":"https://pith.science/api/pith-number/B45EYZM2KKJGX5O4DYXDL4MITP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B45EYZM2KKJGX5O4DYXDL4MITP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B45EYZM2KKJGX5O4DYXDL4MITP/action/storage_attestation","attest_author":"https://pith.science/pith/B45EYZM2KKJGX5O4DYXDL4MITP/action/author_attestation","sign_citation":"https://pith.science/pith/B45EYZM2KKJGX5O4DYXDL4MITP/action/citation_signature","submit_replication":"https://pith.science/pith/B45EYZM2KKJGX5O4DYXDL4MITP/action/replication_record"}},"created_at":"2026-07-05T03:22:06.099420+00:00","updated_at":"2026-07-05T03:22:06.099420+00:00"}