{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:EWUIO5VAQSS7GLDJOKSALL3UTD","short_pith_number":"pith:EWUIO5VA","schema_version":"1.0","canonical_sha256":"25a88776a084a5f32c6972a405af7498c37fa0428485508f2ed7423386548b6e","source":{"kind":"arxiv","id":"1902.02455","version":3},"attestation_state":"computed","paper":{"title":"End-to-end losses based on speaker basis vectors and all-speaker hard negative mining for speaker verification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Ha-Jin Yu, Hee-Soo Heo, Hye-Jin Shim, IL-Ho Yang, Jee-weon Jung, Sung-Hyun Yoon","submitted_at":"2019-02-07T02:55:02Z","abstract_excerpt":"In recent years, speaker verification has primarily performed using deep neural networks that are trained to output embeddings from input features such as spectrograms or Mel-filterbank energies. Studies that design various loss functions, including metric learning have been widely explored. In this study, we propose two end-to-end loss functions for speaker verification using the concept of speaker bases, which are trainable parameters. One loss function is designed to further increase the inter-speaker variation, and the other is designed to conduct the identical concept with hard negative m"},"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":"1902.02455","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2019-02-07T02:55:02Z","cross_cats_sorted":["cs.LG","cs.SD"],"title_canon_sha256":"2f3c1b296f616824de772656956b57b98c27b42a710bc9dd92ba89e9c4ee4e16","abstract_canon_sha256":"a2070b89501ca582970621bca4d83628d2e48fa1dc107688455516060f5cda8b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:40:23.682539Z","signature_b64":"89OSeNFUCeyBSyH9sPetI2IHozfosv1TdOhbK+FxNqKg8EG3uquk02DaJTAVVqDt3MlPGwkaBGmw52jWpMxPDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"25a88776a084a5f32c6972a405af7498c37fa0428485508f2ed7423386548b6e","last_reissued_at":"2026-05-17T23:40:23.681855Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:40:23.681855Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"End-to-end losses based on speaker basis vectors and all-speaker hard negative mining for speaker verification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD"],"primary_cat":"eess.AS","authors_text":"Ha-Jin Yu, Hee-Soo Heo, Hye-Jin Shim, IL-Ho Yang, Jee-weon Jung, Sung-Hyun Yoon","submitted_at":"2019-02-07T02:55:02Z","abstract_excerpt":"In recent years, speaker verification has primarily performed using deep neural networks that are trained to output embeddings from input features such as spectrograms or Mel-filterbank energies. Studies that design various loss functions, including metric learning have been widely explored. In this study, we propose two end-to-end loss functions for speaker verification using the concept of speaker bases, which are trainable parameters. One loss function is designed to further increase the inter-speaker variation, and the other is designed to conduct the identical concept with hard negative m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1902.02455","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"1902.02455","created_at":"2026-05-17T23:40:23.681948+00:00"},{"alias_kind":"arxiv_version","alias_value":"1902.02455v3","created_at":"2026-05-17T23:40:23.681948+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1902.02455","created_at":"2026-05-17T23:40:23.681948+00:00"},{"alias_kind":"pith_short_12","alias_value":"EWUIO5VAQSS7","created_at":"2026-05-18T12:33:15.570797+00:00"},{"alias_kind":"pith_short_16","alias_value":"EWUIO5VAQSS7GLDJ","created_at":"2026-05-18T12:33:15.570797+00:00"},{"alias_kind":"pith_short_8","alias_value":"EWUIO5VA","created_at":"2026-05-18T12:33:15.570797+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EWUIO5VAQSS7GLDJOKSALL3UTD","json":"https://pith.science/pith/EWUIO5VAQSS7GLDJOKSALL3UTD.json","graph_json":"https://pith.science/api/pith-number/EWUIO5VAQSS7GLDJOKSALL3UTD/graph.json","events_json":"https://pith.science/api/pith-number/EWUIO5VAQSS7GLDJOKSALL3UTD/events.json","paper":"https://pith.science/paper/EWUIO5VA"},"agent_actions":{"view_html":"https://pith.science/pith/EWUIO5VAQSS7GLDJOKSALL3UTD","download_json":"https://pith.science/pith/EWUIO5VAQSS7GLDJOKSALL3UTD.json","view_paper":"https://pith.science/paper/EWUIO5VA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1902.02455&json=true","fetch_graph":"https://pith.science/api/pith-number/EWUIO5VAQSS7GLDJOKSALL3UTD/graph.json","fetch_events":"https://pith.science/api/pith-number/EWUIO5VAQSS7GLDJOKSALL3UTD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EWUIO5VAQSS7GLDJOKSALL3UTD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EWUIO5VAQSS7GLDJOKSALL3UTD/action/storage_attestation","attest_author":"https://pith.science/pith/EWUIO5VAQSS7GLDJOKSALL3UTD/action/author_attestation","sign_citation":"https://pith.science/pith/EWUIO5VAQSS7GLDJOKSALL3UTD/action/citation_signature","submit_replication":"https://pith.science/pith/EWUIO5VAQSS7GLDJOKSALL3UTD/action/replication_record"}},"created_at":"2026-05-17T23:40:23.681948+00:00","updated_at":"2026-05-17T23:40:23.681948+00:00"}