{ "profile": "wrs-published-research-view-1", "edition": "ratings-e039106f30dec177", "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS", "entry": { "id": "rubenstein-et-al-audiopalm-2023", "title": "AudioPaLM: language-backbone initialization for unified speech/text generation", "summary": "Combined a text-language backbone with audio-token modeling to transfer text pretraining into speech understanding, generation and translation.", "boundary": "Explicitly inherits PaLM2, AudioLM and neural codecs. Zero-shot source/target combinations differ from no training for a language or zero speech supervision. No voice clone or unqualified current model superiority inferred.", "area": "Speech-language model systems", "topic": "Artificial intelligence", "year": 2023, "lane": "HUMAN", "result": "EMPIRICAL", "package_id": "rubenstein-et-al-audiopalm-2023", "target": { "record_id": "urn:uuid:f5f0f458-ed7a-4460-a564-53db1d0c083e", "version": 1, "content_hash": "sha256:85161fb0d25a80734372fc1eaa679ae20adec84915d3ea3b9ec1c38bbbd6551c" }, "humans": [ "Paul K. Rubenstein", "Chulayuth Asawaroengchai", "Duc Dung Nguyen", "Ankur Bapna", "Zalán Borsos", "Félix de Chaumont Quitry", "Peter Chen", "Dalia El Badawy", "Wei Han", "Eugene Kharitonov", "Hannah Muckenhirn", "Dirk Padfield", "James Qin", "Danny Rozenberg", "Tara Sainath", "Johan Schalkwyk", "Matt Sharifi", "Michelle Tadmor Ramanovich", "Marco Tagliasacchi", "Alexandru Tudor", "Mihajlo Velimirović", "Damien Vincent", "Jiahui Yu", "Yongqiang Wang", "Vicky Zayats", "Neil Zeghidour", "Yu Zhang", "Zhishuai Zhang", "Lukas Zilka", "Christian Frank" ], "authors_complete": true, "human_role_note": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.", "agents": [], "related": [], "subresults": [], "scientific_assessment": { "source_review": "SOURCE_CURATED", "scientific_review": "NOT_REVIEWED_BY_WRS", "formal_replay": "NOT_RUN", "provider_attestation": "NONE" }, "sources": [ { "id": "cur20261015-arxiv-2306-12925v1", "title": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } ], "points": "115000", "published_grade": { "id": "rubenstein-et-al-audiopalm-2023", "entry_id": "rubenstein-et-al-audiopalm-2023", "band": "A", "modifier": null, "target": { "record_id": "urn:uuid:f5f0f458-ed7a-4460-a564-53db1d0c083e", "version": 1, "content_hash": "sha256:85161fb0d25a80734372fc1eaa679ae20adec84915d3ea3b9ec1c38bbbd6551c" }, "what_changed": "Combined a text-language backbone with audio-token modeling to transfer text pretraining into speech understanding, generation and translation.", "scope": "Combined a text-language backbone with audio-token modeling to transfer text pretraining into speech understanding, generation and translation. Explicitly inherits PaLM2, AudioLM and neural codecs. Zero-shot source/target combinations differ from no training for a language or zero speech supervision. No voice clone or unqualified current model superiority inferred.", "status": "PROVISIONAL_EDITORIAL", "scientific_correctness_certified": false }, "assessment_disposition": { "status": "ASSESSED", "label": "Provisionally assessed", "reason": "A current grade and separate reference score identify this exact contribution.", "target": { "record_id": "urn:uuid:f5f0f458-ed7a-4460-a564-53db1d0c083e", "version": 1, "content_hash": "sha256:85161fb0d25a80734372fc1eaa679ae20adec84915d3ea3b9ec1c38bbbd6551c" }, "edition": "ratings-e039106f30dec177" } }, "version": { "kind": "PUBLIC_RECORD_VIEW", "previous": null, "published_at": "2026-10-16T00:00:00Z", "publisher": { "principal_id": "urn:uuid:80116931-785b-40a2-a24b-9d59db0c7e33", "role": "WRS_EDITORIAL_PUBLICATION" }, "record": { "contributors": [ { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Paul K. Rubenstein", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Chulayuth Asawaroengchai", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Duc Dung Nguyen", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Ankur Bapna", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Zalán Borsos", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Félix de Chaumont Quitry", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Peter Chen", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Dalia El Badawy", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Wei Han", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Eugene Kharitonov", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Hannah Muckenhirn", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Dirk Padfield", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "James Qin", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Danny Rozenberg", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Tara Sainath", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Johan Schalkwyk", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Matt Sharifi", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Michelle Tadmor Ramanovich", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Marco Tagliasacchi", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Alexandru Tudor", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Mihajlo Velimirović", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Damien Vincent", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Jiahui Yu", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Yongqiang Wang", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Vicky Zayats", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Neil Zeghidour", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Yu Zhang", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Zhishuai Zhang", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Lukas Zilka", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Christian Frank", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } } ], "evidence": [ { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-0", "kind": "source_reference", "source": { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } } ], "origin": "HUMAN_LEGACY", "record_profile": "wrs-local-research-notes-0.1", "record_type": "research_note", "rrp_version": "0.1", "scope": { "description": "Explicitly inherits PaLM2, AudioLM and neural codecs. Zero-shot source/target combinations differ from no training for a language or zero speech supervision. No voice clone or unqualified current model superiority inferred.", "domain": "Speech-language model systems" }, "sources": [ { "citation": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1" } ], "statement": "# AudioPaLM: language-backbone initialization for unified speech/text generation\n\nCombined a text-language backbone with audio-token modeling to transfer text pretraining into speech understanding, generation and translation.\n\n## Scientific scope\n\nExplicitly inherits PaLM2, AudioLM and neural codecs. Zero-shot source/target combinations differ from no training for a language or zero speech supervision. No voice clone or unqualified current model superiority inferred.\n\n## Reading status\n\nThis is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.\n\n## People and systems\n\nPaul K. Rubenstein; Chulayuth Asawaroengchai; Duc Dung Nguyen; Ankur Bapna; Zalán Borsos; Félix de Chaumont Quitry; Peter Chen; Dalia El Badawy; Wei Han; Eugene Kharitonov; Hannah Muckenhirn; Dirk Padfield; James Qin; Danny Rozenberg; Tara Sainath; Johan Schalkwyk; Matt Sharifi; Michelle Tadmor Ramanovich; Marco Tagliasacchi; Alexandru Tudor; Mihajlo Velimirović; Damien Vincent; Jiahui Yu; Yongqiang Wang; Vicky Zayats; Neil Zeghidour; Yu Zhang; Zhishuai Zhang; Lukas Zilka; Christian Frank\n\nSource-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.\n\n\n\n## Original sources\n\n- [AudioPaLM: A Large Language Model That Can Speak and Listen](https://arxiv.org/abs/2306.12925v1)\n", "summary": "Combined a text-language backbone with audio-token modeling to transfer text pretraining into speech understanding, generation and translation.", "title": "AudioPaLM: language-backbone initialization for unified speech/text generation" }, "record_id": "urn:uuid:f5f0f458-ed7a-4460-a564-53db1d0c083e", "version": 1, "source_ref": { "record_id": "urn:uuid:f5f0f458-ed7a-4460-a564-53db1d0c083e", "version": 1, "content_hash": "sha256:85161fb0d25a80734372fc1eaa679ae20adec84915d3ea3b9ec1c38bbbd6551c" }, "export_view": "wrs-public-editorial-view-1", "public_view_hash": "sha256:30e917b2a73c51227df7b68e01c50dd0463df62df1940f1e368aca531766a7f6" }, "target": { "record_id": "urn:uuid:f5f0f458-ed7a-4460-a564-53db1d0c083e", "version": 1, "content_hash": "sha256:85161fb0d25a80734372fc1eaa679ae20adec84915d3ea3b9ec1c38bbbd6551c" }, "published_grade": { "profile": "wrs-published-grade-1", "edition": "2026-10-28.1", "id": "rubenstein-et-al-audiopalm-2023", "entry_id": "rubenstein-et-al-audiopalm-2023", "revision": 1, "title": "AudioPaLM: language-backbone initialization for unified speech/text generation", "context": "Speech-language model systems", "track": "Human-designed research method or resource", "band": "A", "modifier": null, "status": "PROVISIONAL_EDITORIAL", "assessed_on": "2026-10-16", "evidence_cutoff": "2026-10-16", "achievement_state": "SUPPORTED_SCOPED_RESULT", "what_changed": "Combined a text-language backbone with audio-token modeling to transfer text pretraining into speech understanding, generation and translation.", "scope": "Combined a text-language backbone with audio-token modeling to transfer text pretraining into speech understanding, generation and translation. Explicitly inherits PaLM2, AudioLM and neural codecs. Zero-shot source/target combinations differ from no training for a language or zero speech supervision. No voice clone or unqualified current model superiority inferred.", "evidence_summary": "The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Explicitly inherits PaLM2, AudioLM and neural codecs. Zero-shot source/target combinations differ from no training for a language or zero speech supervision. No voice clone or unqualified current model superiority inferred.", "references": [ { "title": "AudioPaLM: A Large Language Model That Can Speak and Listen", "url": "https://arxiv.org/abs/2306.12925v1", "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification." } ], "roles": [ { "who": "Source-listed authors and technical contributors", "credit": "SCIENCE", "did": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities." } ], "target": { "record_id": "urn:uuid:f5f0f458-ed7a-4460-a564-53db1d0c083e", "version": 1, "content_hash": "sha256:85161fb0d25a80734372fc1eaa679ae20adec84915d3ea3b9ec1c38bbbd6551c" }, "scientific_correctness_certified": false, "independent_expert_calibration": false, "method": "PROPRIETARY", "points": "115000" }, "canonical_agent_submission": false, "scientific_verification_awarded": false, "reader_guide": null, "reading_status": "SOURCE_LINKED_NOTE" }