{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:D65H7VHGHPNKLYVAVWNGGFVONY","short_pith_number":"pith:D65H7VHG","schema_version":"1.0","canonical_sha256":"1fba7fd4e63bdaa5e2a0ad9a6316ae6e0e4b3ee5f92e89d9a99abe3ce2497372","source":{"kind":"arxiv","id":"2110.05968","version":2},"attestation_state":"computed","paper":{"title":"Improving Character Error Rate Is Not Equal to Having Clean Speech: Speech Enhancement for ASR Systems with Black-box Acoustic Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"eess.AS","authors_text":"Ryosuke Sawata, Shusuke Takahashi, Yosuke Kashiwagi","submitted_at":"2021-10-12T12:51:53Z","abstract_excerpt":"A deep neural network (DNN)-based speech enhancement (SE) aiming to maximize the performance of an automatic speech recognition (ASR) system is proposed in this paper. In order to optimize the DNN-based SE model in terms of the character error rate (CER), which is one of the metric to evaluate the ASR system and generally non-differentiable, our method uses two DNNs: one for speech processing and one for mimicking the output CERs derived through an acoustic model (AM). Then both of DNNs are alternately optimized in the training phase. Even if the AM is a black-box, e.g., like one provided by a"},"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":"2110.05968","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2021-10-12T12:51:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2cc211b16643e1ffe11c9d8241329b0886120112271cc02f5b3de059bb3f93b0","abstract_canon_sha256":"28dc2b0886fe80a7322828ddfae2b05193121fd06df857c79e0b27d448da9538"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:58:55.345517Z","signature_b64":"wHVLWZSCgWvG7NEUuaKBkpnT8ztITwGDKt+x0NclWQKonS/PH40tt+ELzgrEmQbCplJ+Ax3kf4fkVtzdB0EJCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1fba7fd4e63bdaa5e2a0ad9a6316ae6e0e4b3ee5f92e89d9a99abe3ce2497372","last_reissued_at":"2026-07-05T03:58:55.345071Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:58:55.345071Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Character Error Rate Is Not Equal to Having Clean Speech: Speech Enhancement for ASR Systems with Black-box Acoustic Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"eess.AS","authors_text":"Ryosuke Sawata, Shusuke Takahashi, Yosuke Kashiwagi","submitted_at":"2021-10-12T12:51:53Z","abstract_excerpt":"A deep neural network (DNN)-based speech enhancement (SE) aiming to maximize the performance of an automatic speech recognition (ASR) system is proposed in this paper. In order to optimize the DNN-based SE model in terms of the character error rate (CER), which is one of the metric to evaluate the ASR system and generally non-differentiable, our method uses two DNNs: one for speech processing and one for mimicking the output CERs derived through an acoustic model (AM). Then both of DNNs are alternately optimized in the training phase. Even if the AM is a black-box, e.g., like one provided by a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.05968","kind":"arxiv","version":2},"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/2110.05968/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":"2110.05968","created_at":"2026-07-05T03:58:55.345127+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.05968v2","created_at":"2026-07-05T03:58:55.345127+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.05968","created_at":"2026-07-05T03:58:55.345127+00:00"},{"alias_kind":"pith_short_12","alias_value":"D65H7VHGHPNK","created_at":"2026-07-05T03:58:55.345127+00:00"},{"alias_kind":"pith_short_16","alias_value":"D65H7VHGHPNKLYVA","created_at":"2026-07-05T03:58:55.345127+00:00"},{"alias_kind":"pith_short_8","alias_value":"D65H7VHG","created_at":"2026-07-05T03:58:55.345127+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/D65H7VHGHPNKLYVAVWNGGFVONY","json":"https://pith.science/pith/D65H7VHGHPNKLYVAVWNGGFVONY.json","graph_json":"https://pith.science/api/pith-number/D65H7VHGHPNKLYVAVWNGGFVONY/graph.json","events_json":"https://pith.science/api/pith-number/D65H7VHGHPNKLYVAVWNGGFVONY/events.json","paper":"https://pith.science/paper/D65H7VHG"},"agent_actions":{"view_html":"https://pith.science/pith/D65H7VHGHPNKLYVAVWNGGFVONY","download_json":"https://pith.science/pith/D65H7VHGHPNKLYVAVWNGGFVONY.json","view_paper":"https://pith.science/paper/D65H7VHG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.05968&json=true","fetch_graph":"https://pith.science/api/pith-number/D65H7VHGHPNKLYVAVWNGGFVONY/graph.json","fetch_events":"https://pith.science/api/pith-number/D65H7VHGHPNKLYVAVWNGGFVONY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/D65H7VHGHPNKLYVAVWNGGFVONY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/D65H7VHGHPNKLYVAVWNGGFVONY/action/storage_attestation","attest_author":"https://pith.science/pith/D65H7VHGHPNKLYVAVWNGGFVONY/action/author_attestation","sign_citation":"https://pith.science/pith/D65H7VHGHPNKLYVAVWNGGFVONY/action/citation_signature","submit_replication":"https://pith.science/pith/D65H7VHGHPNKLYVAVWNGGFVONY/action/replication_record"}},"created_at":"2026-07-05T03:58:55.345127+00:00","updated_at":"2026-07-05T03:58:55.345127+00:00"}