{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:H2DFXHE6HMNTTK7557TUMRHU2E","short_pith_number":"pith:H2DFXHE6","schema_version":"1.0","canonical_sha256":"3e865b9c9e3b1b39abfdefe74644f4d11882a54f6dc2c98ef9db9e2cfbb641f8","source":{"kind":"arxiv","id":"2411.13159","version":1},"attestation_state":"computed","paper":{"title":"Hard-Synth: Synthesizing Diverse Hard Samples for ASR using Zero-Shot TTS and LLM","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Hao Yang, Huan Zhao, Jiawei Yu, Jinsong Su, Min Zhang, Wei Tang, Xiaofeng Zhao, Xiaosong Qiao, Yuang Li","submitted_at":"2024-11-20T09:49:37Z","abstract_excerpt":"Text-to-speech (TTS) models have been widely adopted to enhance automatic speech recognition (ASR) systems using text-only corpora, thereby reducing the cost of labeling real speech data. Existing research primarily utilizes additional text data and predefined speech styles supported by TTS models. In this paper, we propose Hard-Synth, a novel ASR data augmentation method that leverages large language models (LLMs) and advanced zero-shot TTS. Our approach employs LLMs to generate diverse in-domain text through rewriting, without relying on additional text data. Rather than using predefined spe"},"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":"2411.13159","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-20T09:49:37Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"055b121cf04923e5312319cf7e57c0b7d13a3a7c0dfc6e4f39b6831a5b236ee2","abstract_canon_sha256":"0fc9bc7ad372cf1272ee4c423798e6afe659991b0cef2d7d8febeb7ecf7fb287"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:38:06.043021Z","signature_b64":"7zzjm8uCH+NOr5ftVcyFsBios6l6hWzCqT/8sUjmpyfLZ76bi1zBUPXt1UNLPH3uMrFbmGfju/GR1Mkr72tqDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3e865b9c9e3b1b39abfdefe74644f4d11882a54f6dc2c98ef9db9e2cfbb641f8","last_reissued_at":"2026-07-05T09:38:06.042583Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:38:06.042583Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hard-Synth: Synthesizing Diverse Hard Samples for ASR using Zero-Shot TTS and LLM","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Hao Yang, Huan Zhao, Jiawei Yu, Jinsong Su, Min Zhang, Wei Tang, Xiaofeng Zhao, Xiaosong Qiao, Yuang Li","submitted_at":"2024-11-20T09:49:37Z","abstract_excerpt":"Text-to-speech (TTS) models have been widely adopted to enhance automatic speech recognition (ASR) systems using text-only corpora, thereby reducing the cost of labeling real speech data. Existing research primarily utilizes additional text data and predefined speech styles supported by TTS models. In this paper, we propose Hard-Synth, a novel ASR data augmentation method that leverages large language models (LLMs) and advanced zero-shot TTS. Our approach employs LLMs to generate diverse in-domain text through rewriting, without relying on additional text data. Rather than using predefined spe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.13159","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/2411.13159/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":"2411.13159","created_at":"2026-07-05T09:38:06.042636+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.13159v1","created_at":"2026-07-05T09:38:06.042636+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.13159","created_at":"2026-07-05T09:38:06.042636+00:00"},{"alias_kind":"pith_short_12","alias_value":"H2DFXHE6HMNT","created_at":"2026-07-05T09:38:06.042636+00:00"},{"alias_kind":"pith_short_16","alias_value":"H2DFXHE6HMNTTK75","created_at":"2026-07-05T09:38:06.042636+00:00"},{"alias_kind":"pith_short_8","alias_value":"H2DFXHE6","created_at":"2026-07-05T09:38:06.042636+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/H2DFXHE6HMNTTK7557TUMRHU2E","json":"https://pith.science/pith/H2DFXHE6HMNTTK7557TUMRHU2E.json","graph_json":"https://pith.science/api/pith-number/H2DFXHE6HMNTTK7557TUMRHU2E/graph.json","events_json":"https://pith.science/api/pith-number/H2DFXHE6HMNTTK7557TUMRHU2E/events.json","paper":"https://pith.science/paper/H2DFXHE6"},"agent_actions":{"view_html":"https://pith.science/pith/H2DFXHE6HMNTTK7557TUMRHU2E","download_json":"https://pith.science/pith/H2DFXHE6HMNTTK7557TUMRHU2E.json","view_paper":"https://pith.science/paper/H2DFXHE6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.13159&json=true","fetch_graph":"https://pith.science/api/pith-number/H2DFXHE6HMNTTK7557TUMRHU2E/graph.json","fetch_events":"https://pith.science/api/pith-number/H2DFXHE6HMNTTK7557TUMRHU2E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/H2DFXHE6HMNTTK7557TUMRHU2E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/H2DFXHE6HMNTTK7557TUMRHU2E/action/storage_attestation","attest_author":"https://pith.science/pith/H2DFXHE6HMNTTK7557TUMRHU2E/action/author_attestation","sign_citation":"https://pith.science/pith/H2DFXHE6HMNTTK7557TUMRHU2E/action/citation_signature","submit_replication":"https://pith.science/pith/H2DFXHE6HMNTTK7557TUMRHU2E/action/replication_record"}},"created_at":"2026-07-05T09:38:06.042636+00:00","updated_at":"2026-07-05T09:38:06.042636+00:00"}