{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:UZB2AOCEFU4DBEQ33PQFAKCVD6","short_pith_number":"pith:UZB2AOCE","schema_version":"1.0","canonical_sha256":"a643a038442d3830921bdbe05028551f9d3996c17c34ce307175cc6355b0b321","source":{"kind":"arxiv","id":"2303.04255","version":1},"attestation_state":"computed","paper":{"title":"Self-supervised speech representation learning for keyword-spotting with light-weight transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Chenyang Gao, Francesco Caliva, Yue Gu, Yuzong Liu","submitted_at":"2023-03-07T21:54:35Z","abstract_excerpt":"Self-supervised speech representation learning (S3RL) is revolutionizing the way we leverage the ever-growing availability of data. While S3RL related studies typically use large models, we employ light-weight networks to comply with tight memory of compute-constrained devices. We demonstrate the effectiveness of S3RL on a keyword-spotting (KS) problem by using transformers with 330k parameters and propose a mechanism to enhance utterance-wise distinction, which proves crucial for improving performance on classification tasks. On the Google speech commands v2 dataset, the proposed method appli"},"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":"2303.04255","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2023-03-07T21:54:35Z","cross_cats_sorted":["cs.LG","eess.AS"],"title_canon_sha256":"1b94d8afef41d2139d4f8d6c4d4845a6e8b5d4cd204fb2d28cdc60b783280ddb","abstract_canon_sha256":"3b5b6cd57df8e8dadec25706f8fe7998b443178fbd75d5d380ff89cf61b135e2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:49:05.937057Z","signature_b64":"EAxOD1A9LnvgvZUxM0vCJTkScF/RkE24FnYHea4UVwF9tIcMHo7VhYmeA0s6jptBwFR7IjNNcq4rf7T0DAxyDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a643a038442d3830921bdbe05028551f9d3996c17c34ce307175cc6355b0b321","last_reissued_at":"2026-07-05T05:49:05.936615Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:49:05.936615Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-supervised speech representation learning for keyword-spotting with light-weight transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Chenyang Gao, Francesco Caliva, Yue Gu, Yuzong Liu","submitted_at":"2023-03-07T21:54:35Z","abstract_excerpt":"Self-supervised speech representation learning (S3RL) is revolutionizing the way we leverage the ever-growing availability of data. While S3RL related studies typically use large models, we employ light-weight networks to comply with tight memory of compute-constrained devices. We demonstrate the effectiveness of S3RL on a keyword-spotting (KS) problem by using transformers with 330k parameters and propose a mechanism to enhance utterance-wise distinction, which proves crucial for improving performance on classification tasks. On the Google speech commands v2 dataset, the proposed method appli"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.04255","kind":"arxiv","version":1},"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/2303.04255/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":"2303.04255","created_at":"2026-07-05T05:49:05.936676+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.04255v1","created_at":"2026-07-05T05:49:05.936676+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.04255","created_at":"2026-07-05T05:49:05.936676+00:00"},{"alias_kind":"pith_short_12","alias_value":"UZB2AOCEFU4D","created_at":"2026-07-05T05:49:05.936676+00:00"},{"alias_kind":"pith_short_16","alias_value":"UZB2AOCEFU4DBEQ3","created_at":"2026-07-05T05:49:05.936676+00:00"},{"alias_kind":"pith_short_8","alias_value":"UZB2AOCE","created_at":"2026-07-05T05:49:05.936676+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/UZB2AOCEFU4DBEQ33PQFAKCVD6","json":"https://pith.science/pith/UZB2AOCEFU4DBEQ33PQFAKCVD6.json","graph_json":"https://pith.science/api/pith-number/UZB2AOCEFU4DBEQ33PQFAKCVD6/graph.json","events_json":"https://pith.science/api/pith-number/UZB2AOCEFU4DBEQ33PQFAKCVD6/events.json","paper":"https://pith.science/paper/UZB2AOCE"},"agent_actions":{"view_html":"https://pith.science/pith/UZB2AOCEFU4DBEQ33PQFAKCVD6","download_json":"https://pith.science/pith/UZB2AOCEFU4DBEQ33PQFAKCVD6.json","view_paper":"https://pith.science/paper/UZB2AOCE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.04255&json=true","fetch_graph":"https://pith.science/api/pith-number/UZB2AOCEFU4DBEQ33PQFAKCVD6/graph.json","fetch_events":"https://pith.science/api/pith-number/UZB2AOCEFU4DBEQ33PQFAKCVD6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UZB2AOCEFU4DBEQ33PQFAKCVD6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UZB2AOCEFU4DBEQ33PQFAKCVD6/action/storage_attestation","attest_author":"https://pith.science/pith/UZB2AOCEFU4DBEQ33PQFAKCVD6/action/author_attestation","sign_citation":"https://pith.science/pith/UZB2AOCEFU4DBEQ33PQFAKCVD6/action/citation_signature","submit_replication":"https://pith.science/pith/UZB2AOCEFU4DBEQ33PQFAKCVD6/action/replication_record"}},"created_at":"2026-07-05T05:49:05.936676+00:00","updated_at":"2026-07-05T05:49:05.936676+00:00"}