{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:5A6QK6ZCOXHHTNTFC7LHNM6VMW","short_pith_number":"pith:5A6QK6ZC","schema_version":"1.0","canonical_sha256":"e83d057b2275ce79b66517d676b3d565b5a0f8b5a2b9b50084d92f12fa5ed618","source":{"kind":"arxiv","id":"1911.03912","version":3},"attestation_state":"computed","paper":{"title":"Effectiveness of self-supervised pre-training for speech recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Abdelrahman Mohamed, Alexei Baevski, Michael Auli","submitted_at":"2019-11-10T11:50:14Z","abstract_excerpt":"We compare self-supervised representation learning algorithms which either explicitly quantize the audio data or learn representations without quantization. We find the former to be more accurate since it builds a good vocabulary of the data through vq-wav2vec [1] to enable learning of effective representations in subsequent BERT training. Different to previous work, we directly fine-tune the pre-trained BERT models on transcribed speech using a Connectionist Temporal Classification (CTC) loss instead of feeding the representations into a task-specific model. We also propose a BERT-style model"},"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":"1911.03912","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-11-10T11:50:14Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"baff48d913325adefa4aa6676258b3476a060b3d58a0b2c0e2e4939895f3e9ee","abstract_canon_sha256":"46b654e0ac151509ef455f2fed5fd585b3df74dcd4a80f571ab02d67520b396b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:03:41.448775Z","signature_b64":"pdd6/7s/4+NxHswpjCR257qzw9ffDGrRGvVc8Oh1zGBDkKfFAzL1tnb2tCz+adlWph2ucicHKpPeSl8/p875AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e83d057b2275ce79b66517d676b3d565b5a0f8b5a2b9b50084d92f12fa5ed618","last_reissued_at":"2026-07-05T01:03:41.448355Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:03:41.448355Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Effectiveness of self-supervised pre-training for speech recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Abdelrahman Mohamed, Alexei Baevski, Michael Auli","submitted_at":"2019-11-10T11:50:14Z","abstract_excerpt":"We compare self-supervised representation learning algorithms which either explicitly quantize the audio data or learn representations without quantization. We find the former to be more accurate since it builds a good vocabulary of the data through vq-wav2vec [1] to enable learning of effective representations in subsequent BERT training. Different to previous work, we directly fine-tune the pre-trained BERT models on transcribed speech using a Connectionist Temporal Classification (CTC) loss instead of feeding the representations into a task-specific model. We also propose a BERT-style model"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.03912","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/1911.03912/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":"1911.03912","created_at":"2026-07-05T01:03:41.448415+00:00"},{"alias_kind":"arxiv_version","alias_value":"1911.03912v3","created_at":"2026-07-05T01:03:41.448415+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.03912","created_at":"2026-07-05T01:03:41.448415+00:00"},{"alias_kind":"pith_short_12","alias_value":"5A6QK6ZCOXHH","created_at":"2026-07-05T01:03:41.448415+00:00"},{"alias_kind":"pith_short_16","alias_value":"5A6QK6ZCOXHHTNTF","created_at":"2026-07-05T01:03:41.448415+00:00"},{"alias_kind":"pith_short_8","alias_value":"5A6QK6ZC","created_at":"2026-07-05T01:03:41.448415+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.04814","citing_title":"Pitch Accent Detection improves Pretrained Automatic Speech Recognition","ref_index":50,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5A6QK6ZCOXHHTNTFC7LHNM6VMW","json":"https://pith.science/pith/5A6QK6ZCOXHHTNTFC7LHNM6VMW.json","graph_json":"https://pith.science/api/pith-number/5A6QK6ZCOXHHTNTFC7LHNM6VMW/graph.json","events_json":"https://pith.science/api/pith-number/5A6QK6ZCOXHHTNTFC7LHNM6VMW/events.json","paper":"https://pith.science/paper/5A6QK6ZC"},"agent_actions":{"view_html":"https://pith.science/pith/5A6QK6ZCOXHHTNTFC7LHNM6VMW","download_json":"https://pith.science/pith/5A6QK6ZCOXHHTNTFC7LHNM6VMW.json","view_paper":"https://pith.science/paper/5A6QK6ZC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1911.03912&json=true","fetch_graph":"https://pith.science/api/pith-number/5A6QK6ZCOXHHTNTFC7LHNM6VMW/graph.json","fetch_events":"https://pith.science/api/pith-number/5A6QK6ZCOXHHTNTFC7LHNM6VMW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5A6QK6ZCOXHHTNTFC7LHNM6VMW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5A6QK6ZCOXHHTNTFC7LHNM6VMW/action/storage_attestation","attest_author":"https://pith.science/pith/5A6QK6ZCOXHHTNTFC7LHNM6VMW/action/author_attestation","sign_citation":"https://pith.science/pith/5A6QK6ZCOXHHTNTFC7LHNM6VMW/action/citation_signature","submit_replication":"https://pith.science/pith/5A6QK6ZCOXHHTNTFC7LHNM6VMW/action/replication_record"}},"created_at":"2026-07-05T01:03:41.448415+00:00","updated_at":"2026-07-05T01:03:41.448415+00:00"}