{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:V2H7D2BUWNMITL5JOT7OO44J6K","short_pith_number":"pith:V2H7D2BU","schema_version":"1.0","canonical_sha256":"ae8ff1e834b35889afa974fee77389f2a32a6e396ca8f55d1ef31b2ccc4babac","source":{"kind":"arxiv","id":"1905.07149","version":3},"attestation_state":"computed","paper":{"title":"End-to-end Adaptation with Backpropagation through WFST for On-device Speech Recognition System","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.SD"],"primary_cat":"eess.AS","authors_text":"Emiru Tsunoo, Satoshi Asakawa, Toshiyuki Kumakura, Yosuke Kashiwagi","submitted_at":"2019-05-17T07:49:33Z","abstract_excerpt":"An on-device DNN-HMM speech recognition system efficiently works with a limited vocabulary in the presence of a variety of predictable noise. In such a case, vocabulary and environment adaptation is highly effective. In this paper, we propose a novel method of end-to-end (E2E) adaptation, which adjusts not only an acoustic model (AM) but also a weighted finite-state transducer (WFST). We convert a pretrained WFST to a trainable neural network and adapt the system to target environments/vocabulary by E2E joint training with an AM. We replicate Viterbi decoding with forward--backward neural netw"},"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":"1905.07149","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2019-05-17T07:49:33Z","cross_cats_sorted":["cs.CL","cs.SD"],"title_canon_sha256":"3e34a943e3104486ef5ef736ce8e1a86ad2ff167feb84d126f37bc26c91f9273","abstract_canon_sha256":"97fb48e9c6a45fe7b58e5c196cd2b17602cae6dcf83b1add381b8da0f872c471"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:42:41.651668Z","signature_b64":"A3DzNW6mYIpNrbeFYb4qstxZjZcbcBjpMT8VgxtpncIqNjZmUhog1nDfOqUB8dQVyTYX5DHV47qCQ9UQGZlXCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ae8ff1e834b35889afa974fee77389f2a32a6e396ca8f55d1ef31b2ccc4babac","last_reissued_at":"2026-05-17T23:42:41.651032Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:42:41.651032Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"End-to-end Adaptation with Backpropagation through WFST for On-device Speech Recognition System","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.SD"],"primary_cat":"eess.AS","authors_text":"Emiru Tsunoo, Satoshi Asakawa, Toshiyuki Kumakura, Yosuke Kashiwagi","submitted_at":"2019-05-17T07:49:33Z","abstract_excerpt":"An on-device DNN-HMM speech recognition system efficiently works with a limited vocabulary in the presence of a variety of predictable noise. In such a case, vocabulary and environment adaptation is highly effective. In this paper, we propose a novel method of end-to-end (E2E) adaptation, which adjusts not only an acoustic model (AM) but also a weighted finite-state transducer (WFST). We convert a pretrained WFST to a trainable neural network and adapt the system to target environments/vocabulary by E2E joint training with an AM. We replicate Viterbi decoding with forward--backward neural netw"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1905.07149","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":"1905.07149","created_at":"2026-05-17T23:42:41.651123+00:00"},{"alias_kind":"arxiv_version","alias_value":"1905.07149v3","created_at":"2026-05-17T23:42:41.651123+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1905.07149","created_at":"2026-05-17T23:42:41.651123+00:00"},{"alias_kind":"pith_short_12","alias_value":"V2H7D2BUWNMI","created_at":"2026-05-18T12:33:30.264802+00:00"},{"alias_kind":"pith_short_16","alias_value":"V2H7D2BUWNMITL5J","created_at":"2026-05-18T12:33:30.264802+00:00"},{"alias_kind":"pith_short_8","alias_value":"V2H7D2BU","created_at":"2026-05-18T12:33:30.264802+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/V2H7D2BUWNMITL5JOT7OO44J6K","json":"https://pith.science/pith/V2H7D2BUWNMITL5JOT7OO44J6K.json","graph_json":"https://pith.science/api/pith-number/V2H7D2BUWNMITL5JOT7OO44J6K/graph.json","events_json":"https://pith.science/api/pith-number/V2H7D2BUWNMITL5JOT7OO44J6K/events.json","paper":"https://pith.science/paper/V2H7D2BU"},"agent_actions":{"view_html":"https://pith.science/pith/V2H7D2BUWNMITL5JOT7OO44J6K","download_json":"https://pith.science/pith/V2H7D2BUWNMITL5JOT7OO44J6K.json","view_paper":"https://pith.science/paper/V2H7D2BU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1905.07149&json=true","fetch_graph":"https://pith.science/api/pith-number/V2H7D2BUWNMITL5JOT7OO44J6K/graph.json","fetch_events":"https://pith.science/api/pith-number/V2H7D2BUWNMITL5JOT7OO44J6K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V2H7D2BUWNMITL5JOT7OO44J6K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V2H7D2BUWNMITL5JOT7OO44J6K/action/storage_attestation","attest_author":"https://pith.science/pith/V2H7D2BUWNMITL5JOT7OO44J6K/action/author_attestation","sign_citation":"https://pith.science/pith/V2H7D2BUWNMITL5JOT7OO44J6K/action/citation_signature","submit_replication":"https://pith.science/pith/V2H7D2BUWNMITL5JOT7OO44J6K/action/replication_record"}},"created_at":"2026-05-17T23:42:41.651123+00:00","updated_at":"2026-05-17T23:42:41.651123+00:00"}