{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:COVXGQ6IV4I6JG2AP4MHDDLTNS","short_pith_number":"pith:COVXGQ6I","schema_version":"1.0","canonical_sha256":"13ab7343c8af11e49b407f18718d736cbcc2eb230a1a868029f58c456be912c6","source":{"kind":"arxiv","id":"2107.05604","version":2},"attestation_state":"computed","paper":{"title":"Direct speech-to-speech translation with discrete units","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.AS"],"primary_cat":"cs.CL","authors_text":"Adam Polyak, Ann Lee, Changhan Wang, Jiatao Gu, Juan Pino, Peng-Jen Chen, Qing He, Sravya Popuri, Wei-Ning Hsu, Xutai Ma, Yossi Adi, Yun Tang","submitted_at":"2021-07-12T17:40:43Z","abstract_excerpt":"We present a direct speech-to-speech translation (S2ST) model that translates speech from one language to speech in another language without relying on intermediate text generation. We tackle the problem by first applying a self-supervised discrete speech encoder on the target speech and then training a sequence-to-sequence speech-to-unit translation (S2UT) model to predict the discrete representations of the target speech. When target text transcripts are available, we design a joint speech and text training framework that enables the model to generate dual modality output (speech and text) s"},"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":"2107.05604","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-07-12T17:40:43Z","cross_cats_sorted":["cs.LG","eess.AS"],"title_canon_sha256":"1784d45f388a4c89507ae036d40d42c9d1a34a855abadc350d3ee7411886d10f","abstract_canon_sha256":"4b766469ac95f1560cf1b9f3e25696ca78167c193dbda9a48ce1a33586d660b9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:07:04.646844Z","signature_b64":"w42zLHVWtMQ23gfyCK+NyRvz5TEjsxm2GZFg/nZp3qYlU1PtYpBCVVb8WZCgZYAlQ0GHP//hAFHQ1ont2NrHDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"13ab7343c8af11e49b407f18718d736cbcc2eb230a1a868029f58c456be912c6","last_reissued_at":"2026-07-05T04:07:04.646339Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:07:04.646339Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Direct speech-to-speech translation with discrete units","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.AS"],"primary_cat":"cs.CL","authors_text":"Adam Polyak, Ann Lee, Changhan Wang, Jiatao Gu, Juan Pino, Peng-Jen Chen, Qing He, Sravya Popuri, Wei-Ning Hsu, Xutai Ma, Yossi Adi, Yun Tang","submitted_at":"2021-07-12T17:40:43Z","abstract_excerpt":"We present a direct speech-to-speech translation (S2ST) model that translates speech from one language to speech in another language without relying on intermediate text generation. We tackle the problem by first applying a self-supervised discrete speech encoder on the target speech and then training a sequence-to-sequence speech-to-unit translation (S2UT) model to predict the discrete representations of the target speech. When target text transcripts are available, we design a joint speech and text training framework that enables the model to generate dual modality output (speech and text) s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.05604","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/2107.05604/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":"2107.05604","created_at":"2026-07-05T04:07:04.646410+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.05604v2","created_at":"2026-07-05T04:07:04.646410+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.05604","created_at":"2026-07-05T04:07:04.646410+00:00"},{"alias_kind":"pith_short_12","alias_value":"COVXGQ6IV4I6","created_at":"2026-07-05T04:07:04.646410+00:00"},{"alias_kind":"pith_short_16","alias_value":"COVXGQ6IV4I6JG2A","created_at":"2026-07-05T04:07:04.646410+00:00"},{"alias_kind":"pith_short_8","alias_value":"COVXGQ6I","created_at":"2026-07-05T04:07:04.646410+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2210.13438","citing_title":"High Fidelity Neural Audio Compression","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/COVXGQ6IV4I6JG2AP4MHDDLTNS","json":"https://pith.science/pith/COVXGQ6IV4I6JG2AP4MHDDLTNS.json","graph_json":"https://pith.science/api/pith-number/COVXGQ6IV4I6JG2AP4MHDDLTNS/graph.json","events_json":"https://pith.science/api/pith-number/COVXGQ6IV4I6JG2AP4MHDDLTNS/events.json","paper":"https://pith.science/paper/COVXGQ6I"},"agent_actions":{"view_html":"https://pith.science/pith/COVXGQ6IV4I6JG2AP4MHDDLTNS","download_json":"https://pith.science/pith/COVXGQ6IV4I6JG2AP4MHDDLTNS.json","view_paper":"https://pith.science/paper/COVXGQ6I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.05604&json=true","fetch_graph":"https://pith.science/api/pith-number/COVXGQ6IV4I6JG2AP4MHDDLTNS/graph.json","fetch_events":"https://pith.science/api/pith-number/COVXGQ6IV4I6JG2AP4MHDDLTNS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/COVXGQ6IV4I6JG2AP4MHDDLTNS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/COVXGQ6IV4I6JG2AP4MHDDLTNS/action/storage_attestation","attest_author":"https://pith.science/pith/COVXGQ6IV4I6JG2AP4MHDDLTNS/action/author_attestation","sign_citation":"https://pith.science/pith/COVXGQ6IV4I6JG2AP4MHDDLTNS/action/citation_signature","submit_replication":"https://pith.science/pith/COVXGQ6IV4I6JG2AP4MHDDLTNS/action/replication_record"}},"created_at":"2026-07-05T04:07:04.646410+00:00","updated_at":"2026-07-05T04:07:04.646410+00:00"}