{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:54G3ZUZ333LLE7SSK27RO6SBHI","short_pith_number":"pith:54G3ZUZ3","schema_version":"1.0","canonical_sha256":"ef0dbcd33bded6b27e5256bf177a413a310a601e8b270accaf087cd2d97214ca","source":{"kind":"arxiv","id":"1811.03700","version":2},"attestation_state":"computed","paper":{"title":"A Comparison of Lattice-free Discriminative Training Criteria for Purely Sequence-Trained Neural Network Acoustic Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","eess.AS","stat.ML"],"primary_cat":"cs.LG","authors_text":"Chao Weng, Dong Yu","submitted_at":"2018-11-08T22:37:55Z","abstract_excerpt":"In this work, three lattice-free (LF) discriminative training criteria for purely sequence-trained neural network acoustic models are compared on LVCSR tasks, namely maximum mutual information (MMI), boosted maximum mutual information (bMMI) and state-level minimum Bayes risk (sMBR). We demonstrate that, analogous to LF-MMI, a neural network acoustic model can also be trained from scratch using LF-bMMI or LF-sMBR criteria respectively without the need of cross-entropy pre-training. Furthermore, experimental results on Switchboard-300hrs and Switchboard+Fisher-2100hrs datasets show that models "},"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":"1811.03700","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-08T22:37:55Z","cross_cats_sorted":["cs.AI","cs.CL","eess.AS","stat.ML"],"title_canon_sha256":"c47332de6b100b086ed95d70a96c1c2288abf1120c093f70695ca24b32ef35f7","abstract_canon_sha256":"afc57b4148fc539e609a579efeefb5b65f6bc9b9a29ee8698e3ba361c55a2ed3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:59:01.309859Z","signature_b64":"uSdYEy2gZqRhdr6Nwg92PGGFdQA70moBqpHv7SIg5UvImyjKMEubCS+QbRUAy2OMYi+lY9WLbAqf2UESQr04DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ef0dbcd33bded6b27e5256bf177a413a310a601e8b270accaf087cd2d97214ca","last_reissued_at":"2026-05-17T23:59:01.309207Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:59:01.309207Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Comparison of Lattice-free Discriminative Training Criteria for Purely Sequence-Trained Neural Network Acoustic Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","eess.AS","stat.ML"],"primary_cat":"cs.LG","authors_text":"Chao Weng, Dong Yu","submitted_at":"2018-11-08T22:37:55Z","abstract_excerpt":"In this work, three lattice-free (LF) discriminative training criteria for purely sequence-trained neural network acoustic models are compared on LVCSR tasks, namely maximum mutual information (MMI), boosted maximum mutual information (bMMI) and state-level minimum Bayes risk (sMBR). We demonstrate that, analogous to LF-MMI, a neural network acoustic model can also be trained from scratch using LF-bMMI or LF-sMBR criteria respectively without the need of cross-entropy pre-training. Furthermore, experimental results on Switchboard-300hrs and Switchboard+Fisher-2100hrs datasets show that models "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.03700","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":""},"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":"1811.03700","created_at":"2026-05-17T23:59:01.309311+00:00"},{"alias_kind":"arxiv_version","alias_value":"1811.03700v2","created_at":"2026-05-17T23:59:01.309311+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.03700","created_at":"2026-05-17T23:59:01.309311+00:00"},{"alias_kind":"pith_short_12","alias_value":"54G3ZUZ333LL","created_at":"2026-05-18T12:32:05.422762+00:00"},{"alias_kind":"pith_short_16","alias_value":"54G3ZUZ333LLE7SS","created_at":"2026-05-18T12:32:05.422762+00:00"},{"alias_kind":"pith_short_8","alias_value":"54G3ZUZ3","created_at":"2026-05-18T12:32:05.422762+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/54G3ZUZ333LLE7SSK27RO6SBHI","json":"https://pith.science/pith/54G3ZUZ333LLE7SSK27RO6SBHI.json","graph_json":"https://pith.science/api/pith-number/54G3ZUZ333LLE7SSK27RO6SBHI/graph.json","events_json":"https://pith.science/api/pith-number/54G3ZUZ333LLE7SSK27RO6SBHI/events.json","paper":"https://pith.science/paper/54G3ZUZ3"},"agent_actions":{"view_html":"https://pith.science/pith/54G3ZUZ333LLE7SSK27RO6SBHI","download_json":"https://pith.science/pith/54G3ZUZ333LLE7SSK27RO6SBHI.json","view_paper":"https://pith.science/paper/54G3ZUZ3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1811.03700&json=true","fetch_graph":"https://pith.science/api/pith-number/54G3ZUZ333LLE7SSK27RO6SBHI/graph.json","fetch_events":"https://pith.science/api/pith-number/54G3ZUZ333LLE7SSK27RO6SBHI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/54G3ZUZ333LLE7SSK27RO6SBHI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/54G3ZUZ333LLE7SSK27RO6SBHI/action/storage_attestation","attest_author":"https://pith.science/pith/54G3ZUZ333LLE7SSK27RO6SBHI/action/author_attestation","sign_citation":"https://pith.science/pith/54G3ZUZ333LLE7SSK27RO6SBHI/action/citation_signature","submit_replication":"https://pith.science/pith/54G3ZUZ333LLE7SSK27RO6SBHI/action/replication_record"}},"created_at":"2026-05-17T23:59:01.309311+00:00","updated_at":"2026-05-17T23:59:01.309311+00:00"}