{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:4JRQ6I3X5AAIQRYQ5FTPOSKGME","short_pith_number":"pith:4JRQ6I3X","schema_version":"1.0","canonical_sha256":"e2630f2377e800884710e966f74946611adbb3c02920ef97f3efc746a898419c","source":{"kind":"arxiv","id":"2409.16005","version":1},"attestation_state":"computed","paper":{"title":"Bridging Speech and Text: Enhancing ASR with Pinyin-to-Character Pre-training in LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Eng Siong Chng, Peng Yizhou, Xionghu Zhong, Yang Yuhang","submitted_at":"2024-09-24T12:06:31Z","abstract_excerpt":"The integration of large language models (LLMs) with pre-trained speech models has opened up new avenues in automatic speech recognition (ASR). While LLMs excel in multimodal understanding tasks, effectively leveraging their capabilities for ASR remains a significant challenge. This paper presents a novel training approach to enhance LLM performance in ASR tasks. We propose pre-training LLMs on Pinyin embedding sequences, which represent pronunciation features, to generate corresponding Chinese characters. This step enables the LLM to adapt to generating text from pronunciation features before"},"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":"2409.16005","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-09-24T12:06:31Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"e620820a0715a4f31081c2180410ea157cafef1749366d536fe534498399bf0c","abstract_canon_sha256":"d8ebc5fc6bafb41697f8e1e9656069a8966f2ef3e5f4bd522aa2d6347cf78bac"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:11:03.528444Z","signature_b64":"a0UcWZmPR4trETdFoQkftzb2eTflgzHR4zK9ppC3A63ao+7QARl33r2o3iBn5ZGzHHmGu32o/PKet4fH3gP8CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e2630f2377e800884710e966f74946611adbb3c02920ef97f3efc746a898419c","last_reissued_at":"2026-07-05T09:11:03.527936Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:11:03.527936Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bridging Speech and Text: Enhancing ASR with Pinyin-to-Character Pre-training in LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Eng Siong Chng, Peng Yizhou, Xionghu Zhong, Yang Yuhang","submitted_at":"2024-09-24T12:06:31Z","abstract_excerpt":"The integration of large language models (LLMs) with pre-trained speech models has opened up new avenues in automatic speech recognition (ASR). While LLMs excel in multimodal understanding tasks, effectively leveraging their capabilities for ASR remains a significant challenge. This paper presents a novel training approach to enhance LLM performance in ASR tasks. We propose pre-training LLMs on Pinyin embedding sequences, which represent pronunciation features, to generate corresponding Chinese characters. This step enables the LLM to adapt to generating text from pronunciation features before"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.16005","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/2409.16005/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":"2409.16005","created_at":"2026-07-05T09:11:03.527999+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.16005v1","created_at":"2026-07-05T09:11:03.527999+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.16005","created_at":"2026-07-05T09:11:03.527999+00:00"},{"alias_kind":"pith_short_12","alias_value":"4JRQ6I3X5AAI","created_at":"2026-07-05T09:11:03.527999+00:00"},{"alias_kind":"pith_short_16","alias_value":"4JRQ6I3X5AAIQRYQ","created_at":"2026-07-05T09:11:03.527999+00:00"},{"alias_kind":"pith_short_8","alias_value":"4JRQ6I3X","created_at":"2026-07-05T09:11:03.527999+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2504.18938","citing_title":"RAIR: Retrieval-Augmented Iterative Refinement for Chinese Spelling Correction","ref_index":27,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4JRQ6I3X5AAIQRYQ5FTPOSKGME","json":"https://pith.science/pith/4JRQ6I3X5AAIQRYQ5FTPOSKGME.json","graph_json":"https://pith.science/api/pith-number/4JRQ6I3X5AAIQRYQ5FTPOSKGME/graph.json","events_json":"https://pith.science/api/pith-number/4JRQ6I3X5AAIQRYQ5FTPOSKGME/events.json","paper":"https://pith.science/paper/4JRQ6I3X"},"agent_actions":{"view_html":"https://pith.science/pith/4JRQ6I3X5AAIQRYQ5FTPOSKGME","download_json":"https://pith.science/pith/4JRQ6I3X5AAIQRYQ5FTPOSKGME.json","view_paper":"https://pith.science/paper/4JRQ6I3X","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.16005&json=true","fetch_graph":"https://pith.science/api/pith-number/4JRQ6I3X5AAIQRYQ5FTPOSKGME/graph.json","fetch_events":"https://pith.science/api/pith-number/4JRQ6I3X5AAIQRYQ5FTPOSKGME/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4JRQ6I3X5AAIQRYQ5FTPOSKGME/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4JRQ6I3X5AAIQRYQ5FTPOSKGME/action/storage_attestation","attest_author":"https://pith.science/pith/4JRQ6I3X5AAIQRYQ5FTPOSKGME/action/author_attestation","sign_citation":"https://pith.science/pith/4JRQ6I3X5AAIQRYQ5FTPOSKGME/action/citation_signature","submit_replication":"https://pith.science/pith/4JRQ6I3X5AAIQRYQ5FTPOSKGME/action/replication_record"}},"created_at":"2026-07-05T09:11:03.527999+00:00","updated_at":"2026-07-05T09:11:03.527999+00:00"}