{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:3JB5DNQIR3FMCEK67HNOBR4C34","short_pith_number":"pith:3JB5DNQI","schema_version":"1.0","canonical_sha256":"da43d1b6088ecac1115ef9dae0c782df09c200bff05f824886dacd68bce0703a","source":{"kind":"arxiv","id":"1905.13150","version":2},"attestation_state":"computed","paper":{"title":"Lattice-based lightly-supervised acoustic model training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Joachim Fainberg, Ond\\v{r}ej Klejch, Peter Bell, Steve Renals","submitted_at":"2019-05-30T16:12:26Z","abstract_excerpt":"In the broadcast domain there is an abundance of related text data and partial transcriptions, such as closed captions and subtitles. This text data can be used for lightly supervised training, in which text matching the audio is selected using an existing speech recognition model. Current approaches to light supervision typically filter the data based on matching error rates between the transcriptions and biased decoding hypotheses. In contrast, semi-supervised training does not require matching text data, instead generating a hypothesis using a background language model. State-of-the-art sem"},"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":"1905.13150","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-05-30T16:12:26Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"59f0b60da042c59b4b39b05375711044cbfb79703f5cb58285224f47e53880b3","abstract_canon_sha256":"e10917c65d308bb15ef04e5530172cbfaf835c9e8d1d78f529e4833072fe821a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:40:43.217954Z","signature_b64":"uCjhetOnMDS/xqvk3IS9g9zwr287IytarXa5A+776MSosGG/ZAGuIStvhlriVbXlNjKApG0//ggsToVKVs4UAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"da43d1b6088ecac1115ef9dae0c782df09c200bff05f824886dacd68bce0703a","last_reissued_at":"2026-05-17T23:40:43.217366Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:40:43.217366Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Lattice-based lightly-supervised acoustic model training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Joachim Fainberg, Ond\\v{r}ej Klejch, Peter Bell, Steve Renals","submitted_at":"2019-05-30T16:12:26Z","abstract_excerpt":"In the broadcast domain there is an abundance of related text data and partial transcriptions, such as closed captions and subtitles. This text data can be used for lightly supervised training, in which text matching the audio is selected using an existing speech recognition model. Current approaches to light supervision typically filter the data based on matching error rates between the transcriptions and biased decoding hypotheses. In contrast, semi-supervised training does not require matching text data, instead generating a hypothesis using a background language model. State-of-the-art sem"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1905.13150","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":"1905.13150","created_at":"2026-05-17T23:40:43.217449+00:00"},{"alias_kind":"arxiv_version","alias_value":"1905.13150v2","created_at":"2026-05-17T23:40:43.217449+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1905.13150","created_at":"2026-05-17T23:40:43.217449+00:00"},{"alias_kind":"pith_short_12","alias_value":"3JB5DNQIR3FM","created_at":"2026-05-18T12:33:07.085635+00:00"},{"alias_kind":"pith_short_16","alias_value":"3JB5DNQIR3FMCEK6","created_at":"2026-05-18T12:33:07.085635+00:00"},{"alias_kind":"pith_short_8","alias_value":"3JB5DNQI","created_at":"2026-05-18T12:33:07.085635+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/3JB5DNQIR3FMCEK67HNOBR4C34","json":"https://pith.science/pith/3JB5DNQIR3FMCEK67HNOBR4C34.json","graph_json":"https://pith.science/api/pith-number/3JB5DNQIR3FMCEK67HNOBR4C34/graph.json","events_json":"https://pith.science/api/pith-number/3JB5DNQIR3FMCEK67HNOBR4C34/events.json","paper":"https://pith.science/paper/3JB5DNQI"},"agent_actions":{"view_html":"https://pith.science/pith/3JB5DNQIR3FMCEK67HNOBR4C34","download_json":"https://pith.science/pith/3JB5DNQIR3FMCEK67HNOBR4C34.json","view_paper":"https://pith.science/paper/3JB5DNQI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1905.13150&json=true","fetch_graph":"https://pith.science/api/pith-number/3JB5DNQIR3FMCEK67HNOBR4C34/graph.json","fetch_events":"https://pith.science/api/pith-number/3JB5DNQIR3FMCEK67HNOBR4C34/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3JB5DNQIR3FMCEK67HNOBR4C34/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3JB5DNQIR3FMCEK67HNOBR4C34/action/storage_attestation","attest_author":"https://pith.science/pith/3JB5DNQIR3FMCEK67HNOBR4C34/action/author_attestation","sign_citation":"https://pith.science/pith/3JB5DNQIR3FMCEK67HNOBR4C34/action/citation_signature","submit_replication":"https://pith.science/pith/3JB5DNQIR3FMCEK67HNOBR4C34/action/replication_record"}},"created_at":"2026-05-17T23:40:43.217449+00:00","updated_at":"2026-05-17T23:40:43.217449+00:00"}