{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:QHW4BRHGUTTLL6BEF223QVWJNZ","short_pith_number":"pith:QHW4BRHG","canonical_record":{"source":{"id":"2604.23295","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-25T13:18:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4cdc89dc61763383044d7b026a17cbfacec777d430ad4f94fe64cd639cd7b752","abstract_canon_sha256":"9732cc91f0648f1a1967fe14ea052916acc8f44288feee5e52de6e48f7808319"},"schema_version":"1.0"},"canonical_sha256":"81edc0c4e6a4e6b5f8242eb5b856c96e5f436e81c1534277ee88c6e10d3a01a3","source":{"kind":"arxiv","id":"2604.23295","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.23295","created_at":"2026-05-26T02:04:11Z"},{"alias_kind":"arxiv_version","alias_value":"2604.23295v2","created_at":"2026-05-26T02:04:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.23295","created_at":"2026-05-26T02:04:11Z"},{"alias_kind":"pith_short_12","alias_value":"QHW4BRHGUTTL","created_at":"2026-05-26T02:04:11Z"},{"alias_kind":"pith_short_16","alias_value":"QHW4BRHGUTTLL6BE","created_at":"2026-05-26T02:04:11Z"},{"alias_kind":"pith_short_8","alias_value":"QHW4BRHG","created_at":"2026-05-26T02:04:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:QHW4BRHGUTTLL6BEF223QVWJNZ","target":"record","payload":{"canonical_record":{"source":{"id":"2604.23295","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-25T13:18:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4cdc89dc61763383044d7b026a17cbfacec777d430ad4f94fe64cd639cd7b752","abstract_canon_sha256":"9732cc91f0648f1a1967fe14ea052916acc8f44288feee5e52de6e48f7808319"},"schema_version":"1.0"},"canonical_sha256":"81edc0c4e6a4e6b5f8242eb5b856c96e5f436e81c1534277ee88c6e10d3a01a3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T02:04:11.375981Z","signature_b64":"Wvkq6qE5AQC/DX4smFb8zS/3fX6vXdVXfTn9prhL/nRpMbjt75YzUSd6bal+yQNHEce9uK1A44JZ2OuANAE6CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"81edc0c4e6a4e6b5f8242eb5b856c96e5f436e81c1534277ee88c6e10d3a01a3","last_reissued_at":"2026-05-26T02:04:11.375375Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T02:04:11.375375Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.23295","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-26T02:04:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"V6Cjk1rvjOqxd7tzhGoGygMM8U3ZD+HTLqjFnOyXuCZqO9DAx6cv3v50+ihJ2wPBWU4iJhWXe/dcf69BxOF7Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T18:27:33.845471Z"},"content_sha256":"f7ece969ea880d7adb4b90284e87d767effe607081b83c62aeb7d718a10fb4ef","schema_version":"1.0","event_id":"sha256:f7ece969ea880d7adb4b90284e87d767effe607081b83c62aeb7d718a10fb4ef"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:QHW4BRHGUTTLL6BEF223QVWJNZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Human-1 by Josh Talks: A Full-Duplex Conversational Modeling Framework in Hindi using Real-World Conversations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"The first open full-duplex spoken dialogue system for Hindi learns turn-taking and overlaps directly from 26,000 hours of real conversations.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bhaskar Singh, Mahima Manik, Pranav Sharma, Shobhit Banga","submitted_at":"2026-04-25T13:18:40Z","abstract_excerpt":"Full-duplex spoken dialogue systems can model natural conversational behaviours such as interruptions, overlaps, and backchannels, yet such systems remain largely unexplored for Indian languages. We present the first open, reproducible full-duplex spoken dialogue system for Hindi by adapting Moshi, a state-of-the-art duplex speech architecture, using a custom Hindi tokeniser and training on 26,000 hours of real spontaneous conversations collected from 14,695 speakers with separate speaker channels, enabling direct learning of turn-taking and overlap patterns from natural interactions. To suppo"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We present the first open, reproducible full-duplex spoken dialogue system for Hindi by adapting Moshi, a state-of-the-art duplex speech architecture, using a custom Hindi tokeniser and training on 26,000 hours of real spontaneous conversations collected from 14,695 speakers with separate speaker channels, enabling direct learning of turn-taking and overlap patterns from natural interactions.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That reinitializing only the text-vocabulary-dependent parameters while retaining the pre-trained audio components from the English Moshi model will allow the system to effectively learn and generate natural full-duplex conversational behaviors in Hindi from the collected data.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Human-1 is the first open full-duplex spoken dialogue system for Hindi, created by adapting Moshi with a custom tokenizer and training on 26,000 hours of real-world conversations to enable natural interruptions and overlaps.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"The first open full-duplex spoken dialogue system for Hindi learns turn-taking and overlaps directly from 26,000 hours of real conversations.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"49b55187b0d2389e8aeafb81655c06cc9281926d6a6ff32e92cd2f607be74687"},"source":{"id":"2604.23295","kind":"arxiv","version":2},"verdict":{"id":"e71a0f43-0e7b-4851-b472-1c349c5bae89","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T08:11:33.717109Z","strongest_claim":"We present the first open, reproducible full-duplex spoken dialogue system for Hindi by adapting Moshi, a state-of-the-art duplex speech architecture, using a custom Hindi tokeniser and training on 26,000 hours of real spontaneous conversations collected from 14,695 speakers with separate speaker channels, enabling direct learning of turn-taking and overlap patterns from natural interactions.","one_line_summary":"Human-1 is the first open full-duplex spoken dialogue system for Hindi, created by adapting Moshi with a custom tokenizer and training on 26,000 hours of real-world conversations to enable natural interruptions and overlaps.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That reinitializing only the text-vocabulary-dependent parameters while retaining the pre-trained audio components from the English Moshi model will allow the system to effectively learn and generate natural full-duplex conversational behaviors in Hindi from the collected data.","pith_extraction_headline":"The first open full-duplex spoken dialogue system for Hindi learns turn-taking and overlaps directly from 26,000 hours of real conversations."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.23295/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-21T09:35:57.413618Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T23:16:33.509214Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"5bac3955c7ebc1d3e0a51a325c0f30a0fe1ca1cfc724661ec4a9de0ce0ec9e2d"},"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"},"verdict_id":"e71a0f43-0e7b-4851-b472-1c349c5bae89"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-26T02:04:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"e/d93xSGaaCGHfzLx+Qi+wVC3eBDMojcKm8sc66XvPz7VKqVEdp1tWwILOaTWT55yZps76V5fTQ2r7ykKqJ+Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T18:27:33.845980Z"},"content_sha256":"16b8cd997aac873166fd678f788fb114022fe45f4ff1ab5f2c417afd26ec3563","schema_version":"1.0","event_id":"sha256:16b8cd997aac873166fd678f788fb114022fe45f4ff1ab5f2c417afd26ec3563"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QHW4BRHGUTTLL6BEF223QVWJNZ/bundle.json","state_url":"https://pith.science/pith/QHW4BRHGUTTLL6BEF223QVWJNZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QHW4BRHGUTTLL6BEF223QVWJNZ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-30T18:27:33Z","links":{"resolver":"https://pith.science/pith/QHW4BRHGUTTLL6BEF223QVWJNZ","bundle":"https://pith.science/pith/QHW4BRHGUTTLL6BEF223QVWJNZ/bundle.json","state":"https://pith.science/pith/QHW4BRHGUTTLL6BEF223QVWJNZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QHW4BRHGUTTLL6BEF223QVWJNZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:QHW4BRHGUTTLL6BEF223QVWJNZ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"9732cc91f0648f1a1967fe14ea052916acc8f44288feee5e52de6e48f7808319","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-25T13:18:40Z","title_canon_sha256":"4cdc89dc61763383044d7b026a17cbfacec777d430ad4f94fe64cd639cd7b752"},"schema_version":"1.0","source":{"id":"2604.23295","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.23295","created_at":"2026-05-26T02:04:11Z"},{"alias_kind":"arxiv_version","alias_value":"2604.23295v2","created_at":"2026-05-26T02:04:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.23295","created_at":"2026-05-26T02:04:11Z"},{"alias_kind":"pith_short_12","alias_value":"QHW4BRHGUTTL","created_at":"2026-05-26T02:04:11Z"},{"alias_kind":"pith_short_16","alias_value":"QHW4BRHGUTTLL6BE","created_at":"2026-05-26T02:04:11Z"},{"alias_kind":"pith_short_8","alias_value":"QHW4BRHG","created_at":"2026-05-26T02:04:11Z"}],"graph_snapshots":[{"event_id":"sha256:16b8cd997aac873166fd678f788fb114022fe45f4ff1ab5f2c417afd26ec3563","target":"graph","created_at":"2026-05-26T02:04:11Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"We present the first open, reproducible full-duplex spoken dialogue system for Hindi by adapting Moshi, a state-of-the-art duplex speech architecture, using a custom Hindi tokeniser and training on 26,000 hours of real spontaneous conversations collected from 14,695 speakers with separate speaker channels, enabling direct learning of turn-taking and overlap patterns from natural interactions."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That reinitializing only the text-vocabulary-dependent parameters while retaining the pre-trained audio components from the English Moshi model will allow the system to effectively learn and generate natural full-duplex conversational behaviors in Hindi from the collected data."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Human-1 is the first open full-duplex spoken dialogue system for Hindi, created by adapting Moshi with a custom tokenizer and training on 26,000 hours of real-world conversations to enable natural interruptions and overlaps."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"The first open full-duplex spoken dialogue system for Hindi learns turn-taking and overlaps directly from 26,000 hours of real conversations."}],"snapshot_sha256":"49b55187b0d2389e8aeafb81655c06cc9281926d6a6ff32e92cd2f607be74687"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-21T09:35:57.413618Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T23:16:33.509214Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2604.23295/integrity.json","findings":[],"snapshot_sha256":"5bac3955c7ebc1d3e0a51a325c0f30a0fe1ca1cfc724661ec4a9de0ce0ec9e2d","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Full-duplex spoken dialogue systems can model natural conversational behaviours such as interruptions, overlaps, and backchannels, yet such systems remain largely unexplored for Indian languages. We present the first open, reproducible full-duplex spoken dialogue system for Hindi by adapting Moshi, a state-of-the-art duplex speech architecture, using a custom Hindi tokeniser and training on 26,000 hours of real spontaneous conversations collected from 14,695 speakers with separate speaker channels, enabling direct learning of turn-taking and overlap patterns from natural interactions. To suppo","authors_text":"Bhaskar Singh, Mahima Manik, Pranav Sharma, Shobhit Banga","cross_cats":["cs.AI"],"headline":"The first open full-duplex spoken dialogue system for Hindi learns turn-taking and overlaps directly from 26,000 hours of real conversations.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-25T13:18:40Z","title":"Human-1 by Josh Talks: A Full-Duplex Conversational Modeling Framework in Hindi using Real-World Conversations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.23295","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-08T08:11:33.717109Z","id":"e71a0f43-0e7b-4851-b472-1c349c5bae89","model_set":{"reader":"grok-4.3"},"one_line_summary":"Human-1 is the first open full-duplex spoken dialogue system for Hindi, created by adapting Moshi with a custom tokenizer and training on 26,000 hours of real-world conversations to enable natural interruptions and overlaps.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"The first open full-duplex spoken dialogue system for Hindi learns turn-taking and overlaps directly from 26,000 hours of real conversations.","strongest_claim":"We present the first open, reproducible full-duplex spoken dialogue system for Hindi by adapting Moshi, a state-of-the-art duplex speech architecture, using a custom Hindi tokeniser and training on 26,000 hours of real spontaneous conversations collected from 14,695 speakers with separate speaker channels, enabling direct learning of turn-taking and overlap patterns from natural interactions.","weakest_assumption":"That reinitializing only the text-vocabulary-dependent parameters while retaining the pre-trained audio components from the English Moshi model will allow the system to effectively learn and generate natural full-duplex conversational behaviors in Hindi from the collected data."}},"verdict_id":"e71a0f43-0e7b-4851-b472-1c349c5bae89"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:f7ece969ea880d7adb4b90284e87d767effe607081b83c62aeb7d718a10fb4ef","target":"record","created_at":"2026-05-26T02:04:11Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"9732cc91f0648f1a1967fe14ea052916acc8f44288feee5e52de6e48f7808319","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-25T13:18:40Z","title_canon_sha256":"4cdc89dc61763383044d7b026a17cbfacec777d430ad4f94fe64cd639cd7b752"},"schema_version":"1.0","source":{"id":"2604.23295","kind":"arxiv","version":2}},"canonical_sha256":"81edc0c4e6a4e6b5f8242eb5b856c96e5f436e81c1534277ee88c6e10d3a01a3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"81edc0c4e6a4e6b5f8242eb5b856c96e5f436e81c1534277ee88c6e10d3a01a3","first_computed_at":"2026-05-26T02:04:11.375375Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-26T02:04:11.375375Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Wvkq6qE5AQC/DX4smFb8zS/3fX6vXdVXfTn9prhL/nRpMbjt75YzUSd6bal+yQNHEce9uK1A44JZ2OuANAE6CA==","signature_status":"signed_v1","signed_at":"2026-05-26T02:04:11.375981Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.23295","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f7ece969ea880d7adb4b90284e87d767effe607081b83c62aeb7d718a10fb4ef","sha256:16b8cd997aac873166fd678f788fb114022fe45f4ff1ab5f2c417afd26ec3563"],"state_sha256":"09072358ca8f13105b94bb1ce70ef7f42f41dfca1a81fc3d7271421b44738425"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aIpcyNJpgA1Kb+A8zup6ShkibWW1z2OApNno/ZFBSOQ+oYPEsE9cueUSYsSnRwtJmUu7jCTGpQ7LhggdIyvUBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-30T18:27:33.848371Z","bundle_sha256":"999a25f57a2dabb0cc17c43bb5857dca1f0689946ceba5acf753bf9ed14e69ff"}}