{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.23813", "kind": "paper", "url": "https://arxiv.org/abs/2607.23813"}], "authors": ["Denglin Jiang", "Haoran Zhou", "Anshul Wadhawan", "Brendan Fahy", "Vinay Ramesh", "David Weisberg", "Dmitriy Derkachevskiy", "Helen Sheehan", "Srivas Prasad", "Michele Franceschini"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.23813", "source_url": "https://arxiv.org/abs/2607.23813"}, "categories": ["Finance", "Speech & Audio", "cs.CL"], "description": {"en": "Earnings25 is a finance-domain benchmark for evaluating automatic speech recognition (ASR) on English-language earnings calls. It includes testset-full (498 hours of S&P 500 earnings calls from Q4 2025) and testset-segmented (46 hours of 290 segments from 2025 U.S. earnings calls), with aligned transcripts and metadata like speaker roles and industry labels, enabling speaker- and industry-aware evaluation beyond WER."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.23813", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.23813", "title": "Earnings25"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Bloomberg (United States)"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2607.23813", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2607.23813", "source_url": "https://openalex.org/W7171485952"}, "key": "claire-radar:2607.23813", "modality": null, "name": "Earnings25", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "unknown", "origin_record_id": "bm_earnings25_937fc49b", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "646e58d2fab84680f69baa5f98c8f478360eaefcd74da52b2c6dede09b8ff16b", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2607.23813"}, "publisher": null, "released": "2026-07-26", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.23813", "source_url": "https://arxiv.org/abs/2607.23813"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-23813", "source": "claire_radar", "source_benchmark_id": "2607.23813", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Mixed", "applicationDomains": ["Finance"], "area": "Speech & Audio", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 0, "hfPaperUrl": "https://huggingface.co/papers/2607.23813", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}}, "status": "fresh"}, "capabilities": [], "confidence": "High", "construction": "Aggregate Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "e37fc638ca9da3a66fa076f8aa9dabfc3e80673d78ab7a1750908710a039dc6f", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Earnings25 is a finance-domain benchmark for evaluating automatic speech recognition (ASR) on English-language earnings calls. It includes testset-full (498 hours of S&P 500 earnings calls from Q4 2025) and testset-segmented (46 hours of 290 segments from 2025 U.S. earnings calls), with aligned transcripts and metadata like speaker roles and industry labels, enabling speaker- and industry-aware evaluation beyond WER.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence"], "snippet": "We introduce Earnings25, a finance-domain benchmark for evaluating automatic speech recognition (ASR) on English-language earnings calls under realistic conditions."}, "familyId": "bmf_c5e53bcda5ed", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_earnings25_937fc49b", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Financial Services"], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2607.23813", "paper": "https://arxiv.org/abs/2607.23813", "pdf": "https://arxiv.org/pdf/2607.23813", "project": null, "report": "https://arxiv.org/abs/2607.23813"}, "metrics": [], "motivation": "We introduce Earnings25, a finance-domain benchmark for evaluating automatic speech recognition (ASR) on English-language earnings calls under realistic conditions.", "name": "Earnings25", "oneLine": "Earnings25 is a finance-domain benchmark for evaluating automatic speech recognition (ASR) on English-language earnings calls. It includes testset-full (498 hours of S&P 500 earnings calls from Q4 2025) and testset-segmented (46 hours of 290 segments from 2025 U.S. earnings calls), with aligned transcripts and metadata like speaker roles and industry labels, enabling speaker- and industry-aware evaluation beyond WER.", "paperTitle": "Earnings25: A Comprehensive 500-Hour Speech Benchmark for Finance", "primaryDomain": "Finance", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2607.23813", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "Earnings25 team", "organizationType": "benchmark-organization", "role": "benchmark-publisher", "sourceUrl": "https://arxiv.org/abs/2607.23813"}], "ranking": {"90d": {"confidence": "Low", "coverage": 0.15, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 90}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.0, "value": 0}}, "confidence": "Low", "coverage": 0.15, "rank": null, "score": 43}, "method": "attention-ranking-v13", "rank": null, "score": 43}}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-07-26", "source": {"authors": ["Denglin Jiang", "Haoran Zhou", "Anshul Wadhawan", "Brendan Fahy", "Vinay Ramesh", "David Weisberg", "Dmitriy Derkachevskiy", "Helen Sheehan", "Srivas Prasad", "Michele Franceschini"], "categories": ["cs.CL", "cs.AI"], "id": "2607.23813", "title": "Earnings25: A Comprehensive 500-Hour Speech Benchmark for Finance", "type": "arxiv", "url": "https://arxiv.org/abs/2607.23813"}, "sourceUpdatedAt": "2026-07-26T19:24:00Z", "topics": ["cs.CL"], "venueAttempts": [], "whyItMatters": "Existing ASR benchmarks lack domain-specific, large-scale, and realistic finance data, and typically report only aggregate word error rate. Earnings25 provides a reproducible protocol and structured metadata to support more granular evaluation of ASR systems in the finance domain."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
