{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:TDTP2VQESTI54RHTJLTX4CKYNG","short_pith_number":"pith:TDTP2VQE","schema_version":"1.0","canonical_sha256":"98e6fd560494d1de44f34ae77e095869bdae332db89f92f7c6486ed4b28e9050","source":{"kind":"arxiv","id":"2102.01192","version":2},"attestation_state":"computed","paper":{"title":"Generative Spoken Language Modeling from Raw Audio","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Adam Polyak, Adelrahman Mohamed, Alexei Baevski, Benjamin Bolte, Emmanuel Dupoux, Evgeny Kharitonov, Jade Copet, Kushal Lakhotia, Tu-Anh Nguyen, Wei-Ning Hsu, Yossi Adi","submitted_at":"2021-02-01T21:41:40Z","abstract_excerpt":"We introduce Generative Spoken Language Modeling, the task of learning the acoustic and linguistic characteristics of a language from raw audio (no text, no labels), and a set of metrics to automatically evaluate the learned representations at acoustic and linguistic levels for both encoding and generation. We set up baseline systems consisting of a discrete speech encoder (returning pseudo-text units), a generative language model (trained on pseudo-text), and a speech decoder (generating a waveform from pseudo-text) all trained without supervision and validate the proposed metrics with human "},"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":"2102.01192","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-02-01T21:41:40Z","cross_cats_sorted":[],"title_canon_sha256":"c40b879189175e510355546864ead8c976dacdd548fdad96789c2f32998c69a8","abstract_canon_sha256":"ff72b678067d954a8a12b6c5297430f07d79e578ad7546381ba90f8437ab29f8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:13:02.017906Z","signature_b64":"dtKQBAwCJ6Om9sb6X2hM3RPN9zynPVqaVqG9NsT5imuGUuL9LNZhlXPipJf5IBXMaPw1wMwV+DL6uyQb8i1jCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"98e6fd560494d1de44f34ae77e095869bdae332db89f92f7c6486ed4b28e9050","last_reissued_at":"2026-07-05T03:13:02.017449Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:13:02.017449Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generative Spoken Language Modeling from Raw Audio","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Adam Polyak, Adelrahman Mohamed, Alexei Baevski, Benjamin Bolte, Emmanuel Dupoux, Evgeny Kharitonov, Jade Copet, Kushal Lakhotia, Tu-Anh Nguyen, Wei-Ning Hsu, Yossi Adi","submitted_at":"2021-02-01T21:41:40Z","abstract_excerpt":"We introduce Generative Spoken Language Modeling, the task of learning the acoustic and linguistic characteristics of a language from raw audio (no text, no labels), and a set of metrics to automatically evaluate the learned representations at acoustic and linguistic levels for both encoding and generation. We set up baseline systems consisting of a discrete speech encoder (returning pseudo-text units), a generative language model (trained on pseudo-text), and a speech decoder (generating a waveform from pseudo-text) all trained without supervision and validate the proposed metrics with human "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2102.01192","kind":"arxiv","version":2},"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/2102.01192/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":"2102.01192","created_at":"2026-07-05T03:13:02.017508+00:00"},{"alias_kind":"arxiv_version","alias_value":"2102.01192v2","created_at":"2026-07-05T03:13:02.017508+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.01192","created_at":"2026-07-05T03:13:02.017508+00:00"},{"alias_kind":"pith_short_12","alias_value":"TDTP2VQESTI5","created_at":"2026-07-05T03:13:02.017508+00:00"},{"alias_kind":"pith_short_16","alias_value":"TDTP2VQESTI54RHT","created_at":"2026-07-05T03:13:02.017508+00:00"},{"alias_kind":"pith_short_8","alias_value":"TDTP2VQE","created_at":"2026-07-05T03:13:02.017508+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19910","citing_title":"Light-weight Pronunciation Assessment via Discrete Speech Token Surprisal","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2301.02111","citing_title":"Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TDTP2VQESTI54RHTJLTX4CKYNG","json":"https://pith.science/pith/TDTP2VQESTI54RHTJLTX4CKYNG.json","graph_json":"https://pith.science/api/pith-number/TDTP2VQESTI54RHTJLTX4CKYNG/graph.json","events_json":"https://pith.science/api/pith-number/TDTP2VQESTI54RHTJLTX4CKYNG/events.json","paper":"https://pith.science/paper/TDTP2VQE"},"agent_actions":{"view_html":"https://pith.science/pith/TDTP2VQESTI54RHTJLTX4CKYNG","download_json":"https://pith.science/pith/TDTP2VQESTI54RHTJLTX4CKYNG.json","view_paper":"https://pith.science/paper/TDTP2VQE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2102.01192&json=true","fetch_graph":"https://pith.science/api/pith-number/TDTP2VQESTI54RHTJLTX4CKYNG/graph.json","fetch_events":"https://pith.science/api/pith-number/TDTP2VQESTI54RHTJLTX4CKYNG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TDTP2VQESTI54RHTJLTX4CKYNG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TDTP2VQESTI54RHTJLTX4CKYNG/action/storage_attestation","attest_author":"https://pith.science/pith/TDTP2VQESTI54RHTJLTX4CKYNG/action/author_attestation","sign_citation":"https://pith.science/pith/TDTP2VQESTI54RHTJLTX4CKYNG/action/citation_signature","submit_replication":"https://pith.science/pith/TDTP2VQESTI54RHTJLTX4CKYNG/action/replication_record"}},"created_at":"2026-07-05T03:13:02.017508+00:00","updated_at":"2026-07-05T03:13:02.017508+00:00"}