{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:S6SYX5SMNF4SXJDI6Q7L46EWCM","short_pith_number":"pith:S6SYX5SM","schema_version":"1.0","canonical_sha256":"97a58bf64c69792ba468f43ebe7896131d43c025a2e5358859c871e78ec1456c","source":{"kind":"arxiv","id":"2010.08642","version":1},"attestation_state":"computed","paper":{"title":"Multimodal Speech Recognition with Unstructured Audio Masking","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Desmond Elliott, Florian Metze, Ramon Sanabria, Tejas Srinivasan","submitted_at":"2020-10-16T21:49:20Z","abstract_excerpt":"Visual context has been shown to be useful for automatic speech recognition (ASR) systems when the speech signal is noisy or corrupted. Previous work, however, has only demonstrated the utility of visual context in an unrealistic setting, where a fixed set of words are systematically masked in the audio. In this paper, we simulate a more realistic masking scenario during model training, called RandWordMask, where the masking can occur for any word segment. Our experiments on the Flickr 8K Audio Captions Corpus show that multimodal ASR can generalize to recover different types of masked words i"},"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":"2010.08642","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-16T21:49:20Z","cross_cats_sorted":[],"title_canon_sha256":"3def21ed9abc497dcf1fc97cf119b28efc8a07febb06e13288db7d75277dbbb1","abstract_canon_sha256":"209900621d8e8cac52b254c0b470758ad4d3850802ece23dcb88124514a1bd09"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:43:44.151185Z","signature_b64":"MTtg5dwrq82dxYNafBiJiU7dBZ/Cwhiy6LNf8KgNBolJkqLKCBOObBobAHoYnJ04LWH8tIYRshlNIaNCEEVcCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"97a58bf64c69792ba468f43ebe7896131d43c025a2e5358859c871e78ec1456c","last_reissued_at":"2026-07-05T01:43:44.150811Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:43:44.150811Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multimodal Speech Recognition with Unstructured Audio Masking","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Desmond Elliott, Florian Metze, Ramon Sanabria, Tejas Srinivasan","submitted_at":"2020-10-16T21:49:20Z","abstract_excerpt":"Visual context has been shown to be useful for automatic speech recognition (ASR) systems when the speech signal is noisy or corrupted. Previous work, however, has only demonstrated the utility of visual context in an unrealistic setting, where a fixed set of words are systematically masked in the audio. In this paper, we simulate a more realistic masking scenario during model training, called RandWordMask, where the masking can occur for any word segment. Our experiments on the Flickr 8K Audio Captions Corpus show that multimodal ASR can generalize to recover different types of masked words i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.08642","kind":"arxiv","version":1},"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/2010.08642/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":"2010.08642","created_at":"2026-07-05T01:43:44.150865+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.08642v1","created_at":"2026-07-05T01:43:44.150865+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.08642","created_at":"2026-07-05T01:43:44.150865+00:00"},{"alias_kind":"pith_short_12","alias_value":"S6SYX5SMNF4S","created_at":"2026-07-05T01:43:44.150865+00:00"},{"alias_kind":"pith_short_16","alias_value":"S6SYX5SMNF4SXJDI","created_at":"2026-07-05T01:43:44.150865+00:00"},{"alias_kind":"pith_short_8","alias_value":"S6SYX5SM","created_at":"2026-07-05T01:43:44.150865+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/S6SYX5SMNF4SXJDI6Q7L46EWCM","json":"https://pith.science/pith/S6SYX5SMNF4SXJDI6Q7L46EWCM.json","graph_json":"https://pith.science/api/pith-number/S6SYX5SMNF4SXJDI6Q7L46EWCM/graph.json","events_json":"https://pith.science/api/pith-number/S6SYX5SMNF4SXJDI6Q7L46EWCM/events.json","paper":"https://pith.science/paper/S6SYX5SM"},"agent_actions":{"view_html":"https://pith.science/pith/S6SYX5SMNF4SXJDI6Q7L46EWCM","download_json":"https://pith.science/pith/S6SYX5SMNF4SXJDI6Q7L46EWCM.json","view_paper":"https://pith.science/paper/S6SYX5SM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.08642&json=true","fetch_graph":"https://pith.science/api/pith-number/S6SYX5SMNF4SXJDI6Q7L46EWCM/graph.json","fetch_events":"https://pith.science/api/pith-number/S6SYX5SMNF4SXJDI6Q7L46EWCM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/S6SYX5SMNF4SXJDI6Q7L46EWCM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/S6SYX5SMNF4SXJDI6Q7L46EWCM/action/storage_attestation","attest_author":"https://pith.science/pith/S6SYX5SMNF4SXJDI6Q7L46EWCM/action/author_attestation","sign_citation":"https://pith.science/pith/S6SYX5SMNF4SXJDI6Q7L46EWCM/action/citation_signature","submit_replication":"https://pith.science/pith/S6SYX5SMNF4SXJDI6Q7L46EWCM/action/replication_record"}},"created_at":"2026-07-05T01:43:44.150865+00:00","updated_at":"2026-07-05T01:43:44.150865+00:00"}