{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2609.09672", "kind": "paper", "url": "https://arxiv.org/abs/2609.09672"}, {"kind": "website", "url": "https://zwenyu.github.io/SEA-SpeechBench/"}], "authors": ["Jingyi Liao", "Wenyu Zhang", "Zhuohan Liu", "Yingxu He", "Geyu Lin", "Xunlong Zou", "Shuo Sun", "Syed Ali Redha Alsagoff", "Ai Ti Aw"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2609.09672", "source_url": "https://arxiv.org/abs/2609.09672"}, "categories": ["Factuality", "General AI", "Speech & Audio", "cs.CL"], "description": {"en": "Evaluates speech understanding across 11 Southeast Asian languages via 97,194 samples covering speech processing, paralinguistic analysis, and temporal understanding tasks."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2609.09672", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2609.09672", "title": "SEA-SpeechBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2609.09672", "modality": null, "name": "SEA-SpeechBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"admission_policy_version": "2026-09-09.1", "confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "true", "origin_record_id": "bm_sea-speechbench_336fc1dd", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "68a4e55f33c7612837a635dbe664d4807ea50dcec7e107f5a8dd8e820157f1cc", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-11T00:52:53.785364Z", "source_url": "https://arxiv.org/abs/2609.09672"}, "publisher": null, "released": "2026-09-09", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2609.09672", "source_url": "https://arxiv.org/abs/2609.09672"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2609-09672", "source": "claire_radar", "source_benchmark_id": "2609.09672", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Speech & Audio", "attention": {"asOf": "2026-09-10", "attemptedAt": "2026-09-10T12:24:43.080049Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2609.09672", "observedAt": "2026-09-09T12:31:07.558041Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-09-10T12:24:43.080049Z", "state": "unavailable"}}, "status": "unavailable"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "Broad multilingual coverage and large scale attract interest from speech and language model communities, though the field is moderately crowded.", "score": 52}, "capabilities": ["Factuality"], "confidence": "High", "construction": "Transform Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-11T00:52:53.785364Z", "inputHash": "358db410d3ac9a7c7d58b71b34e1a495135f0a2932f23f547a49c808a983173f", "model": "deepseek-v4-pro"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "We introduce SEA-SpeechBench, to the best of our knowledge, the first large-scale multitask benchmark that evaluates speech understanding in 11 SEA languages", "canonicalNameSource": "abstract", "decisionReason": "Formally named benchmark with clear tasks and scoring, and a public project page enabling reuse.", "evaluationEvidence": {"reuse": "The SEA-SpeechBench benchmark is available at https://zwenyu.github.io/SEA-SpeechBench/", "scoring": "Evaluation of leading open-source and proprietary systems reveals marked performance gaps. Across all models, performance remains underwhelming on temporal understanding, emotion recognition, and speech translation. Prompting in low-resource languages such as Burmese and Tamil lags behind English by up to 41 percentage points.", "task": "9 diverse tasks across 3 categories: speech processing (automatic speech recognition, speech translation, spoken question answering), paralinguistic analysis (emotion, gender, age, speaker recognition), and temporal understanding"}, "model": "deepseek-v4-pro", "reviewedAt": "2026-09-11T00:52:53.785364Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates speech understanding across 11 Southeast Asian languages via 97,194 samples covering speech processing, paralinguistic analysis, and temporal understanding tasks.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "We introduce SEA-SpeechBench, to the best of our knowledge, the first large-scale multitask benchmark that evaluates speech understanding in 11 SEA languages through 97,194 samples across 99 evaluation sets and 597 hours of curated audio data."}, "familyId": "bmf_b81cfe4fc631", "firstSeenAt": "2026-09-10", "heat": null, "id": "bm_sea-speechbench_336fc1dd", "indexedAt": "2026-09-10T00:47:26.889471Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2609.09672", "pdf": "https://arxiv.org/pdf/2609.09672", "project": "https://zwenyu.github.io/SEA-SpeechBench/", "report": "https://arxiv.org/abs/2609.09672"}, "metrics": [], "motivation": "The rapid advancement of audio and multimodal large language models has unlocked transformative speech understanding capabilities, yet evaluation frameworks remain predominantly English-centric, leaving Southeast Asian (SEA) languages critically underrepresented.", "name": "SEA-SpeechBench", "oneLine": "Evaluates speech understanding across 11 Southeast Asian languages via 97,194 samples covering speech processing, paralinguistic analysis, and temporal understanding tasks.", "paperTitle": "SEA-SpeechBench: A Large-Scale Multitask Benchmark for Speech Understanding Across Southeast Asia", "primaryDomain": "General AI", "publication": {"evidence": "Accepted to EMNLP 2026", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2609.09672", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "EMNLP 2026", "verifiedAt": "2026-09-10T00:47:26.889471Z"}, "publications": [], "publishers": [{"name": "SEA-SpeechBench Team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://zwenyu.github.io/SEA-SpeechBench/"}], "ranking": {"30d": {"confidence": "Low", "coverage": 0.0, "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": 30}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": null}, "method": "attention-ranking-v13", "rank": null, "score": null}, "90d": {"confidence": "Low", "coverage": 0.0, "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": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": null}, "method": "attention-ranking-v13", "rank": null, "score": null}, "today": {"confidence": "Low", "coverage": 0.0, "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": 1, "historyDate": "2026-09-09", "rank": null, "score": null, "windowDays": 1}, "level": {"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}, "llmAttentionForecast": {"maxContribution": 0.25, "percentile": 0.62, "role": "bonus", "value": 62}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": 16}, "method": "attention-ranking-v13", "rank": null, "score": 16}}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-09-09", "source": {"authors": ["Jingyi Liao", "Wenyu Zhang", "Zhuohan Liu", "Yingxu He", "Geyu Lin", "Xunlong Zou", "Shuo Sun", "Syed Ali Redha Alsagoff", "Ai Ti Aw"], "categories": ["cs.CL"], "id": "2609.09672", "title": "SEA-SpeechBench: A Large-Scale Multitask Benchmark for Speech Understanding Across Southeast Asia", "type": "arxiv", "url": "https://arxiv.org/abs/2609.09672"}, "sourceUpdatedAt": "2026-09-09T03:39:18Z", "topics": ["cs.CL"], "venueAttempts": [{"decisionRaw": "Accepted to EMNLP 2026", "evidence": [{"level": "author-claim", "observedAt": "2026-09-10T00:47:26.889471Z", "rawValue": "Accepted to EMNLP 2026", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2609.09672"}], "reviewStatus": "accepted", "venueName": "EMNLP 2026"}], "whyItMatters": "Addresses the English-centric bias in speech evaluation by providing a large-scale multilingual benchmark, exposing performance gaps in low-resource SEA languages and guiding inclusive model development."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
