{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:OJCSFRRBCLWO3F67J5DMHZ27C2","short_pith_number":"pith:OJCSFRRB","schema_version":"1.0","canonical_sha256":"724522c62112eced97df4f46c3e75f1690c108aee5051de981614c141ced1b60","source":{"kind":"arxiv","id":"2012.15404","version":1},"attestation_state":"computed","paper":{"title":"Unified Mandarin TTS Front-end Based on Distilled BERT Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SD","authors_text":"Liqun Deng, Yang Zhang, Yasheng Wang","submitted_at":"2020-12-31T02:34:57Z","abstract_excerpt":"The front-end module in a typical Mandarin text-to-speech system (TTS) is composed of a long pipeline of text processing components, which requires extensive efforts to build and is prone to large accumulative model size and cascade errors. In this paper, a pre-trained language model (PLM) based model is proposed to simultaneously tackle the two most important tasks in TTS front-end, i.e., prosodic structure prediction (PSP) and grapheme-to-phoneme (G2P) conversion. We use a pre-trained Chinese BERT[1] as the text encoder and employ multi-task learning technique to adapt it to the two TTS fron"},"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":"2012.15404","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2020-12-31T02:34:57Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"a2126950c649c640ddc0314970e182701168f3b5ffebf0b3cd3b0b1cde1be7b5","abstract_canon_sha256":"b3f1f66e8e79e0afbf09c5164173ce90b492287012183bf29131e8c4a8b0bdfc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:03:18.128201Z","signature_b64":"hf+cU/eeFtYIaZuykSAxUy2nayurQYwskn7N6G6UicdWmfSrmqXixf/fxMEnSyQiP4iETS4+Iq4vgFbE6qypAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"724522c62112eced97df4f46c3e75f1690c108aee5051de981614c141ced1b60","last_reissued_at":"2026-07-05T02:03:18.127801Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:03:18.127801Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unified Mandarin TTS Front-end Based on Distilled BERT Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SD","authors_text":"Liqun Deng, Yang Zhang, Yasheng Wang","submitted_at":"2020-12-31T02:34:57Z","abstract_excerpt":"The front-end module in a typical Mandarin text-to-speech system (TTS) is composed of a long pipeline of text processing components, which requires extensive efforts to build and is prone to large accumulative model size and cascade errors. In this paper, a pre-trained language model (PLM) based model is proposed to simultaneously tackle the two most important tasks in TTS front-end, i.e., prosodic structure prediction (PSP) and grapheme-to-phoneme (G2P) conversion. We use a pre-trained Chinese BERT[1] as the text encoder and employ multi-task learning technique to adapt it to the two TTS fron"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.15404","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/2012.15404/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":"2012.15404","created_at":"2026-07-05T02:03:18.127855+00:00"},{"alias_kind":"arxiv_version","alias_value":"2012.15404v1","created_at":"2026-07-05T02:03:18.127855+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.15404","created_at":"2026-07-05T02:03:18.127855+00:00"},{"alias_kind":"pith_short_12","alias_value":"OJCSFRRBCLWO","created_at":"2026-07-05T02:03:18.127855+00:00"},{"alias_kind":"pith_short_16","alias_value":"OJCSFRRBCLWO3F67","created_at":"2026-07-05T02:03:18.127855+00:00"},{"alias_kind":"pith_short_8","alias_value":"OJCSFRRB","created_at":"2026-07-05T02:03:18.127855+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.07646","citing_title":"Transcript-Prompted Whisper with Dictionary-Enhanced Decoding for Japanese Speech Annotation","ref_index":13,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OJCSFRRBCLWO3F67J5DMHZ27C2","json":"https://pith.science/pith/OJCSFRRBCLWO3F67J5DMHZ27C2.json","graph_json":"https://pith.science/api/pith-number/OJCSFRRBCLWO3F67J5DMHZ27C2/graph.json","events_json":"https://pith.science/api/pith-number/OJCSFRRBCLWO3F67J5DMHZ27C2/events.json","paper":"https://pith.science/paper/OJCSFRRB"},"agent_actions":{"view_html":"https://pith.science/pith/OJCSFRRBCLWO3F67J5DMHZ27C2","download_json":"https://pith.science/pith/OJCSFRRBCLWO3F67J5DMHZ27C2.json","view_paper":"https://pith.science/paper/OJCSFRRB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2012.15404&json=true","fetch_graph":"https://pith.science/api/pith-number/OJCSFRRBCLWO3F67J5DMHZ27C2/graph.json","fetch_events":"https://pith.science/api/pith-number/OJCSFRRBCLWO3F67J5DMHZ27C2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OJCSFRRBCLWO3F67J5DMHZ27C2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OJCSFRRBCLWO3F67J5DMHZ27C2/action/storage_attestation","attest_author":"https://pith.science/pith/OJCSFRRBCLWO3F67J5DMHZ27C2/action/author_attestation","sign_citation":"https://pith.science/pith/OJCSFRRBCLWO3F67J5DMHZ27C2/action/citation_signature","submit_replication":"https://pith.science/pith/OJCSFRRBCLWO3F67J5DMHZ27C2/action/replication_record"}},"created_at":"2026-07-05T02:03:18.127855+00:00","updated_at":"2026-07-05T02:03:18.127855+00:00"}