{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PN3FT5ZERLV2UJDMYAFJPI53AB","short_pith_number":"pith:PN3FT5ZE","schema_version":"1.0","canonical_sha256":"7b7659f7248aebaa246cc00a97a3bb0047dcbf4647a9fc5dde7ff74733a14610","source":{"kind":"arxiv","id":"2505.15333","version":1},"attestation_state":"computed","paper":{"title":"Leveraging Unit Language Guidance to Advance Speech Modeling in Textless Speech-to-Speech Translation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Benyou Wang, Jingbo Zhu, Kaiqi Kou, Peizhuo Liu, Tong Xiao, Weiqiao Shan, Xiangnan Ma, Yuhao Zhang, Yuxin Huang, Zhengtao Yu","submitted_at":"2025-05-21T10:05:25Z","abstract_excerpt":"The success of building textless speech-to-speech translation (S2ST) models has attracted much attention. However, S2ST still faces two main challenges: 1) extracting linguistic features for various speech signals, called cross-modal (CM), and 2) learning alignment of difference languages in long sequences, called cross-lingual (CL). We propose the unit language to overcome the two modeling challenges. The unit language can be considered a text-like representation format, constructed using $n$-gram language modeling. We implement multi-task learning to utilize the unit language in guiding the "},"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":"2505.15333","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-21T10:05:25Z","cross_cats_sorted":["cs.AI","cs.SD","eess.AS"],"title_canon_sha256":"64fcb3b4d219e0827cfdda3fa689fd914d2bd6c0c88ad91b79f59f52dded60da","abstract_canon_sha256":"e3e4638c20aea34cafcb27ceda7b1665916f800dfc580a08a3bfe977b25d258f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:43.939273Z","signature_b64":"0xzOsBh/Tauvz0U9+wShdJYoGhiLnCh2Sjh0hnADM26k+1BOoYXDQTJB1l1Ge4zcbTlWr11nx/iSpHcocnYwBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7b7659f7248aebaa246cc00a97a3bb0047dcbf4647a9fc5dde7ff74733a14610","last_reissued_at":"2026-07-05T11:06:43.938798Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:43.938798Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Leveraging Unit Language Guidance to Advance Speech Modeling in Textless Speech-to-Speech Translation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Benyou Wang, Jingbo Zhu, Kaiqi Kou, Peizhuo Liu, Tong Xiao, Weiqiao Shan, Xiangnan Ma, Yuhao Zhang, Yuxin Huang, Zhengtao Yu","submitted_at":"2025-05-21T10:05:25Z","abstract_excerpt":"The success of building textless speech-to-speech translation (S2ST) models has attracted much attention. However, S2ST still faces two main challenges: 1) extracting linguistic features for various speech signals, called cross-modal (CM), and 2) learning alignment of difference languages in long sequences, called cross-lingual (CL). We propose the unit language to overcome the two modeling challenges. The unit language can be considered a text-like representation format, constructed using $n$-gram language modeling. We implement multi-task learning to utilize the unit language in guiding the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.15333","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/2505.15333/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":"2505.15333","created_at":"2026-07-05T11:06:43.938856+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.15333v1","created_at":"2026-07-05T11:06:43.938856+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.15333","created_at":"2026-07-05T11:06:43.938856+00:00"},{"alias_kind":"pith_short_12","alias_value":"PN3FT5ZERLV2","created_at":"2026-07-05T11:06:43.938856+00:00"},{"alias_kind":"pith_short_16","alias_value":"PN3FT5ZERLV2UJDM","created_at":"2026-07-05T11:06:43.938856+00:00"},{"alias_kind":"pith_short_8","alias_value":"PN3FT5ZE","created_at":"2026-07-05T11:06:43.938856+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/PN3FT5ZERLV2UJDMYAFJPI53AB","json":"https://pith.science/pith/PN3FT5ZERLV2UJDMYAFJPI53AB.json","graph_json":"https://pith.science/api/pith-number/PN3FT5ZERLV2UJDMYAFJPI53AB/graph.json","events_json":"https://pith.science/api/pith-number/PN3FT5ZERLV2UJDMYAFJPI53AB/events.json","paper":"https://pith.science/paper/PN3FT5ZE"},"agent_actions":{"view_html":"https://pith.science/pith/PN3FT5ZERLV2UJDMYAFJPI53AB","download_json":"https://pith.science/pith/PN3FT5ZERLV2UJDMYAFJPI53AB.json","view_paper":"https://pith.science/paper/PN3FT5ZE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.15333&json=true","fetch_graph":"https://pith.science/api/pith-number/PN3FT5ZERLV2UJDMYAFJPI53AB/graph.json","fetch_events":"https://pith.science/api/pith-number/PN3FT5ZERLV2UJDMYAFJPI53AB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PN3FT5ZERLV2UJDMYAFJPI53AB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PN3FT5ZERLV2UJDMYAFJPI53AB/action/storage_attestation","attest_author":"https://pith.science/pith/PN3FT5ZERLV2UJDMYAFJPI53AB/action/author_attestation","sign_citation":"https://pith.science/pith/PN3FT5ZERLV2UJDMYAFJPI53AB/action/citation_signature","submit_replication":"https://pith.science/pith/PN3FT5ZERLV2UJDMYAFJPI53AB/action/replication_record"}},"created_at":"2026-07-05T11:06:43.938856+00:00","updated_at":"2026-07-05T11:06:43.938856+00:00"}