{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2605.26978", "kind": "paper", "url": "https://arxiv.org/abs/2605.26978"}], "authors": ["Hanif Rahman"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2605.26978", "source_url": "https://arxiv.org/abs/2605.26978"}, "categories": ["General AI", "Speech & Audio", "cs.CL"], "description": {"en": "A dated automated screening benchmark for Pashto text-to-speech systems covering completion, transcription error, script fidelity, and audio-language identification."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2605.26978", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2605.26978", "title": "PashtoTTS-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2605.26978", "modality": null, "name": "PashtoTTS-Bench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"admission_policy_version": "2026-09-09.1", "confidence": "Medium", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-09-25T04:40:22Z", "data_status": "primary-source-indexed", "display_eligible": "true", "origin_record_id": "bm_pashtotts-bench_b6d60a8f", "origin_source": "arxiv", "recognition_confidence": "0.65", "record_sha256": "ec005e326f8f930b9570a34449bc97e01a1da973f74035661134ccafba1e35d0", "relation": "introduces", "review_model": "codex-cli-default", "review_state": "ai-reviewed", "reviewed_at": "2026-09-24T17:38:02.578792Z", "source_url": "https://arxiv.org/abs/2605.26978"}, "publisher": null, "released": "2026-05-26", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2605.26978", "source_url": "https://arxiv.org/abs/2605.26978"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2605-26978", "source": "claire_radar", "source_benchmark_id": "2605.26978", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Machine Generated", "applicationDomains": ["General AI"], "area": "Speech & Audio", "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "The benchmark addresses an under-served language and includes reusable system-extension scripts and detailed evaluation artifacts.", "score": 3}, "capabilities": [], "confidence": "Medium", "construction": "Original Synthetic", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-24T17:38:02.578792Z", "inputHash": "d0468dfad84a22fa2ce050c6c06810d7c98976091e71bd3a58604efd900fd864", "model": "codex-cli-default"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "PashtoTTS-Bench: automated screening for low-resource non-Latin-script text-to-speech", "canonicalNameSource": "paper_title", "decisionReason": "The supplied evidence names the benchmark, defines its task and scoring dimensions, and describes a release with scripts for evaluating additional systems.", "evaluationEvidence": {"reuse": "The release provides provider metadata, per-sentence scores, LID audits, failure logs, and scripts for adding systems.", "scoring": "synthesis completion, ASR WER/CER, transcript Script Fidelity Rate, and audio language identification.", "task": "We instantiate INSV-A as PashtoTTS-Bench, a dated benchmark for Pashto TTS."}, "model": "codex-cli-default", "reviewedAt": "2026-09-24T17:38:02.578792Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "A dated automated screening benchmark for Pashto text-to-speech systems covering completion, transcription error, script fidelity, and audio-language identification.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "evaluation protocol evidence"], "snippet": "We instantiate INSV-A as PashtoTTS-Bench, a dated benchmark for Pashto TTS."}, "familyId": "bmf_86d4eba07386", "firstSeenAt": "2026-08-19", "heat": null, "id": "bm_pashtotts-bench_b6d60a8f", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2605.26978", "pdf": "https://arxiv.org/pdf/2605.26978", "project": null, "report": "https://arxiv.org/abs/2605.26978"}, "metrics": [], "motivation": "Text-to-speech (TTS) evaluation for low-resource non-Latin-script languages can fail when it relies on a single ASR round-trip word error rate (WER).", "name": "PashtoTTS-Bench", "oneLine": "A dated automated screening benchmark for Pashto text-to-speech systems covering completion, transcription error, script fidelity, and audio-language identification.", "paperTitle": "PashtoTTS-Bench: automated screening for low-resource non-Latin-script text-to-speech", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2605.26978", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-19T11:05:13.395391Z"}, "publications": [], "readiness": "Paper only", "recognitionConfidence": 0.65, "relation": "introduces", "releasedAt": "2026-05-26", "reviewLifecycle": {"state": "pending"}, "source": {"authors": ["Hanif Rahman"], "categories": ["cs.CL", "cs.SD"], "id": "2605.26978", "title": "PashtoTTS-Bench: automated screening for low-resource non-Latin-script text-to-speech", "type": "arxiv", "url": "https://arxiv.org/abs/2605.26978"}, "sourceUpdatedAt": "2026-05-26T13:03:20Z", "topics": ["cs.CL"], "venueAttempts": [], "whyItMatters": "It separates several failure modes that a single ASR round-trip error score can conceal in low-resource TTS evaluation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
