{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:SXHYIQ2VJVSUNAEAZ5NSOFOWV3","short_pith_number":"pith:SXHYIQ2V","schema_version":"1.0","canonical_sha256":"95cf8443554d65468080cf5b2715d6aecc50bfb4bd71135db7e02b17a5cd4ce7","source":{"kind":"arxiv","id":"2005.07143","version":3},"attestation_state":"computed","paper":{"title":"ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Brecht Desplanques, Jenthe Thienpondt, Kris Demuynck","submitted_at":"2020-05-14T17:02:15Z","abstract_excerpt":"Current speaker verification techniques rely on a neural network to extract speaker representations. The successful x-vector architecture is a Time Delay Neural Network (TDNN) that applies statistics pooling to project variable-length utterances into fixed-length speaker characterizing embeddings. In this paper, we propose multiple enhancements to this architecture based on recent trends in the related fields of face verification and computer vision. Firstly, the initial frame layers can be restructured into 1-dimensional Res2Net modules with impactful skip connections. Similarly to SE-ResNet,"},"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":"2005.07143","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2020-05-14T17:02:15Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"24e046ebba8ea4b5241cbab6c41aed64316f05e83fa4d82fe483ca4903b38bd5","abstract_canon_sha256":"342f637790e8fd5d2758b1375363ec1746452e7a8d5cbe66d296419419113af4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:48:09.480412Z","signature_b64":"FGWqElndmwqG9IFL0lwnbMVCiel/dMub9hgyJEyUTnvAsppvn2FLIBWrJQGG25eERJW0WPLOVcCZlVTddilzBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"95cf8443554d65468080cf5b2715d6aecc50bfb4bd71135db7e02b17a5cd4ce7","last_reissued_at":"2026-07-05T01:48:09.479952Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:48:09.479952Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Brecht Desplanques, Jenthe Thienpondt, Kris Demuynck","submitted_at":"2020-05-14T17:02:15Z","abstract_excerpt":"Current speaker verification techniques rely on a neural network to extract speaker representations. The successful x-vector architecture is a Time Delay Neural Network (TDNN) that applies statistics pooling to project variable-length utterances into fixed-length speaker characterizing embeddings. In this paper, we propose multiple enhancements to this architecture based on recent trends in the related fields of face verification and computer vision. Firstly, the initial frame layers can be restructured into 1-dimensional Res2Net modules with impactful skip connections. Similarly to SE-ResNet,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.07143","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2005.07143/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":"2005.07143","created_at":"2026-07-05T01:48:09.480006+00:00"},{"alias_kind":"arxiv_version","alias_value":"2005.07143v3","created_at":"2026-07-05T01:48:09.480006+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.07143","created_at":"2026-07-05T01:48:09.480006+00:00"},{"alias_kind":"pith_short_12","alias_value":"SXHYIQ2VJVSU","created_at":"2026-07-05T01:48:09.480006+00:00"},{"alias_kind":"pith_short_16","alias_value":"SXHYIQ2VJVSUNAEA","created_at":"2026-07-05T01:48:09.480006+00:00"},{"alias_kind":"pith_short_8","alias_value":"SXHYIQ2V","created_at":"2026-07-05T01:48:09.480006+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":14,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2607.05276","citing_title":"ProPS: Prompted Profile Synthesis for Natural Language-Conditioned Speaker Embedding Distributions","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2606.09831","citing_title":"AI-Driven Analytics of Team-Teaching Talk: Acoustic Patterns across Experience, Cohorts and the Learning Design","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2606.21305","citing_title":"LISE : Listenable Interpretable Speaker Embeddings","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19629","citing_title":"RIVET: Robust Idempotent Voice Attribute Editing","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01729","citing_title":"DRL-CLBA: A Clean Label Backdoor Attack for Speech Classification via DDPG Reinforcement Learning","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09019","citing_title":"TLDR: Compressing Audio Tokens for Efficient Autoregressive Text-to-Speech","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02724","citing_title":"AVTrack: Audio-Visual Tracking in Human-centric Complex Scenes","ref_index":59,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03384","citing_title":"DECKER: Domain-invariant Embedding for Cross-Keyboard Extraction and Recognition","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25991","citing_title":"Fuzzy PyTorch: Rapid Numerical Variability Evaluation for Deep Learning Models","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2506.12606","citing_title":"An Exploration of Mamba for Speech Self-Supervised Models","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2509.15113","citing_title":"Low-rank surrogate modeling and stochastic zero-order optimization for training of neural networks with black-box layers","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19679","citing_title":"MMControl: Unified Multi-Modal Control for Joint Audio-Video Generation","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05683","citing_title":"Time-Domain Voice Identity Morphing (TD-VIM): A Signal-Level Approach to Morphing Attacks on Speaker Verification Systems","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13229","citing_title":"ProSDD: Learning Prosodic Representations for Speech Deepfake Detection against Expressive and Emotional Attacks","ref_index":50,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SXHYIQ2VJVSUNAEAZ5NSOFOWV3","json":"https://pith.science/pith/SXHYIQ2VJVSUNAEAZ5NSOFOWV3.json","graph_json":"https://pith.science/api/pith-number/SXHYIQ2VJVSUNAEAZ5NSOFOWV3/graph.json","events_json":"https://pith.science/api/pith-number/SXHYIQ2VJVSUNAEAZ5NSOFOWV3/events.json","paper":"https://pith.science/paper/SXHYIQ2V"},"agent_actions":{"view_html":"https://pith.science/pith/SXHYIQ2VJVSUNAEAZ5NSOFOWV3","download_json":"https://pith.science/pith/SXHYIQ2VJVSUNAEAZ5NSOFOWV3.json","view_paper":"https://pith.science/paper/SXHYIQ2V","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2005.07143&json=true","fetch_graph":"https://pith.science/api/pith-number/SXHYIQ2VJVSUNAEAZ5NSOFOWV3/graph.json","fetch_events":"https://pith.science/api/pith-number/SXHYIQ2VJVSUNAEAZ5NSOFOWV3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SXHYIQ2VJVSUNAEAZ5NSOFOWV3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SXHYIQ2VJVSUNAEAZ5NSOFOWV3/action/storage_attestation","attest_author":"https://pith.science/pith/SXHYIQ2VJVSUNAEAZ5NSOFOWV3/action/author_attestation","sign_citation":"https://pith.science/pith/SXHYIQ2VJVSUNAEAZ5NSOFOWV3/action/citation_signature","submit_replication":"https://pith.science/pith/SXHYIQ2VJVSUNAEAZ5NSOFOWV3/action/replication_record"}},"created_at":"2026-07-05T01:48:09.480006+00:00","updated_at":"2026-07-05T01:48:09.480006+00:00"}