{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:VVFT6YGXS4XCZCDDBOG7AG52TW","short_pith_number":"pith:VVFT6YGX","schema_version":"1.0","canonical_sha256":"ad4b3f60d7972e2c88630b8df01bba9da7d12d91d47802982024139d3a680a57","source":{"kind":"arxiv","id":"2203.15591","version":1},"attestation_state":"computed","paper":{"title":"Earnings-22: A Practical Benchmark for Accents in the Wild","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Corey Miller, Miguel del Rio, Peter Ha, Quinten McNamara, Shipra Chandra","submitted_at":"2022-03-29T14:02:57Z","abstract_excerpt":"Modern automatic speech recognition (ASR) systems have achieved superhuman Word Error Rate (WER) on many common corpora despite lacking adequate performance on speech in the wild. Beyond that, there is a lack of real-world, accented corpora to properly benchmark academic and commercial models. To ensure this type of speech is represented in ASR benchmarking, we present Earnings-22, a 125 file, 119 hour corpus of English-language earnings calls gathered from global companies. We run a comparison across 4 commercial models showing the variation in performance when taking country of origin into c"},"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":"2203.15591","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2022-03-29T14:02:57Z","cross_cats_sorted":[],"title_canon_sha256":"e0621c65806dd718383dcaa9e2905d084130651988fdd2e54a1a136bbe74cbf1","abstract_canon_sha256":"3b4716b281120c73a7f5b5989dcc58b2afc8503f77c99afaef5c2ccb528efb19"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:09:27.683428Z","signature_b64":"7ahwAQFRxv2Zfag+ySb6x8pzolcy/5KF/lA21+uDQtfXaXVuKlZDQABMfSWld+EcTW9cJ5kgtJUd3s6WIaRZBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ad4b3f60d7972e2c88630b8df01bba9da7d12d91d47802982024139d3a680a57","last_reissued_at":"2026-07-05T04:09:27.682982Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:09:27.682982Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Earnings-22: A Practical Benchmark for Accents in the Wild","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Corey Miller, Miguel del Rio, Peter Ha, Quinten McNamara, Shipra Chandra","submitted_at":"2022-03-29T14:02:57Z","abstract_excerpt":"Modern automatic speech recognition (ASR) systems have achieved superhuman Word Error Rate (WER) on many common corpora despite lacking adequate performance on speech in the wild. Beyond that, there is a lack of real-world, accented corpora to properly benchmark academic and commercial models. To ensure this type of speech is represented in ASR benchmarking, we present Earnings-22, a 125 file, 119 hour corpus of English-language earnings calls gathered from global companies. We run a comparison across 4 commercial models showing the variation in performance when taking country of origin into c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.15591","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/2203.15591/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":"2203.15591","created_at":"2026-07-05T04:09:27.683046+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.15591v1","created_at":"2026-07-05T04:09:27.683046+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.15591","created_at":"2026-07-05T04:09:27.683046+00:00"},{"alias_kind":"pith_short_12","alias_value":"VVFT6YGXS4XC","created_at":"2026-07-05T04:09:27.683046+00:00"},{"alias_kind":"pith_short_16","alias_value":"VVFT6YGXS4XCZCDD","created_at":"2026-07-05T04:09:27.683046+00:00"},{"alias_kind":"pith_short_8","alias_value":"VVFT6YGX","created_at":"2026-07-05T04:09:27.683046+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23048","citing_title":"HALAS: A Human-Annotated Dataset of Hallucinations of Modern ASR Systems","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08486","citing_title":"TRADE: Transducer-Augmented Decoder for Speech LLM","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07354","citing_title":"Contextual Earnings-22: A Speech Recognition Benchmark with Custom Vocabulary in the Wild","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27543","citing_title":"AppTek Call-Center Dialogues: A Multi-Accent Long-Form Benchmark for English ASR","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14604","citing_title":"Hijacking Large Audio-Language Models via Context-Agnostic and Imperceptible Auditory Prompt Injection","ref_index":72,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VVFT6YGXS4XCZCDDBOG7AG52TW","json":"https://pith.science/pith/VVFT6YGXS4XCZCDDBOG7AG52TW.json","graph_json":"https://pith.science/api/pith-number/VVFT6YGXS4XCZCDDBOG7AG52TW/graph.json","events_json":"https://pith.science/api/pith-number/VVFT6YGXS4XCZCDDBOG7AG52TW/events.json","paper":"https://pith.science/paper/VVFT6YGX"},"agent_actions":{"view_html":"https://pith.science/pith/VVFT6YGXS4XCZCDDBOG7AG52TW","download_json":"https://pith.science/pith/VVFT6YGXS4XCZCDDBOG7AG52TW.json","view_paper":"https://pith.science/paper/VVFT6YGX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.15591&json=true","fetch_graph":"https://pith.science/api/pith-number/VVFT6YGXS4XCZCDDBOG7AG52TW/graph.json","fetch_events":"https://pith.science/api/pith-number/VVFT6YGXS4XCZCDDBOG7AG52TW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VVFT6YGXS4XCZCDDBOG7AG52TW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VVFT6YGXS4XCZCDDBOG7AG52TW/action/storage_attestation","attest_author":"https://pith.science/pith/VVFT6YGXS4XCZCDDBOG7AG52TW/action/author_attestation","sign_citation":"https://pith.science/pith/VVFT6YGXS4XCZCDDBOG7AG52TW/action/citation_signature","submit_replication":"https://pith.science/pith/VVFT6YGXS4XCZCDDBOG7AG52TW/action/replication_record"}},"created_at":"2026-07-05T04:09:27.683046+00:00","updated_at":"2026-07-05T04:09:27.683046+00:00"}