{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:S44QER6UWABQY3GQ5R32CGWJAP","short_pith_number":"pith:S44QER6U","schema_version":"1.0","canonical_sha256":"97390247d4b0030c6cd0ec77a11ac903ffbf97ddc53b9af34abcb26bcf4ea143","source":{"kind":"arxiv","id":"2107.08661","version":5},"attestation_state":"computed","paper":{"title":"Translatotron 2: High-quality direct speech-to-speech translation with voice preservation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Michelle Tadmor Ramanovich, Roi Pomerantz, Tal Remez, Ye Jia","submitted_at":"2021-07-19T07:43:49Z","abstract_excerpt":"We present Translatotron 2, a neural direct speech-to-speech translation model that can be trained end-to-end. Translatotron 2 consists of a speech encoder, a linguistic decoder, an acoustic synthesizer, and a single attention module that connects them together. Experimental results on three datasets consistently show that Translatotron 2 outperforms the original Translatotron by a large margin on both translation quality (up to +15.5 BLEU) and speech generation quality, and approaches the same of cascade systems. In addition, we propose a simple method for preserving speakers' voices from 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":"2107.08661","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-07-19T07:43:49Z","cross_cats_sorted":["cs.LG","cs.SD","eess.AS"],"title_canon_sha256":"dda3c09106dce87a127e6da6ce7faf84e85d02540e11cb0e601a70c750b46af4","abstract_canon_sha256":"50897b9808a0d9b9c432e6ebe90887da8f9c573dae849a7c020b7480edc46655"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:24:22.332441Z","signature_b64":"MQjtCdIcfFFnWEtaPjTqP790bQBfu8/iCR4PpwSSs4OXbyArVyb1oMnWRUlH0a44F0wPMSjWcV7CyDf5lQo6Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"97390247d4b0030c6cd0ec77a11ac903ffbf97ddc53b9af34abcb26bcf4ea143","last_reissued_at":"2026-07-05T04:24:22.331985Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:24:22.331985Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Translatotron 2: High-quality direct speech-to-speech translation with voice preservation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Michelle Tadmor Ramanovich, Roi Pomerantz, Tal Remez, Ye Jia","submitted_at":"2021-07-19T07:43:49Z","abstract_excerpt":"We present Translatotron 2, a neural direct speech-to-speech translation model that can be trained end-to-end. Translatotron 2 consists of a speech encoder, a linguistic decoder, an acoustic synthesizer, and a single attention module that connects them together. Experimental results on three datasets consistently show that Translatotron 2 outperforms the original Translatotron by a large margin on both translation quality (up to +15.5 BLEU) and speech generation quality, and approaches the same of cascade systems. In addition, we propose a simple method for preserving speakers' voices from the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.08661","kind":"arxiv","version":5},"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/2107.08661/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":"2107.08661","created_at":"2026-07-05T04:24:22.332040+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.08661v5","created_at":"2026-07-05T04:24:22.332040+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.08661","created_at":"2026-07-05T04:24:22.332040+00:00"},{"alias_kind":"pith_short_12","alias_value":"S44QER6UWABQ","created_at":"2026-07-05T04:24:22.332040+00:00"},{"alias_kind":"pith_short_16","alias_value":"S44QER6UWABQY3GQ","created_at":"2026-07-05T04:24:22.332040+00:00"},{"alias_kind":"pith_short_8","alias_value":"S44QER6U","created_at":"2026-07-05T04:24:22.332040+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/S44QER6UWABQY3GQ5R32CGWJAP","json":"https://pith.science/pith/S44QER6UWABQY3GQ5R32CGWJAP.json","graph_json":"https://pith.science/api/pith-number/S44QER6UWABQY3GQ5R32CGWJAP/graph.json","events_json":"https://pith.science/api/pith-number/S44QER6UWABQY3GQ5R32CGWJAP/events.json","paper":"https://pith.science/paper/S44QER6U"},"agent_actions":{"view_html":"https://pith.science/pith/S44QER6UWABQY3GQ5R32CGWJAP","download_json":"https://pith.science/pith/S44QER6UWABQY3GQ5R32CGWJAP.json","view_paper":"https://pith.science/paper/S44QER6U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.08661&json=true","fetch_graph":"https://pith.science/api/pith-number/S44QER6UWABQY3GQ5R32CGWJAP/graph.json","fetch_events":"https://pith.science/api/pith-number/S44QER6UWABQY3GQ5R32CGWJAP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/S44QER6UWABQY3GQ5R32CGWJAP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/S44QER6UWABQY3GQ5R32CGWJAP/action/storage_attestation","attest_author":"https://pith.science/pith/S44QER6UWABQY3GQ5R32CGWJAP/action/author_attestation","sign_citation":"https://pith.science/pith/S44QER6UWABQY3GQ5R32CGWJAP/action/citation_signature","submit_replication":"https://pith.science/pith/S44QER6UWABQY3GQ5R32CGWJAP/action/replication_record"}},"created_at":"2026-07-05T04:24:22.332040+00:00","updated_at":"2026-07-05T04:24:22.332040+00:00"}