{ "schema_version": 1, "builder_plan": { "scenario": "Your workload might fit a smaller or differently structured model.", "change": "Check for available weights and a compatible runtime before planning a product trial.", "baseline": "Your current model on identical held-out tasks and the same resource budget.", "steps": [ "Locate the paper's implementation and check its license, model support, data needs, and hardware requirements. Stop here if you cannot obtain the required artifacts.", "Save a small set of representative product tasks and expected outcomes. Keep evaluation tasks out of any training or tuning, and choose quality and cost limits before running.", "Run the baseline and the proposed change on identical inputs. Keep other settings fixed and repeat paired runs to expose variation.", "Save per-task outputs, failures, timings, and all usage costs. Review regressions before expanding the test." ], "measure": "Task success, latency, memory, context limits, and migration cost.", "decision": "Keep the change only if it meets your preselected quality and latency limits and improves the metric you care about after setup and operating costs. Otherwise retain the baseline. This is an exploration plan, not a validated recommendation." }, "builder_review": null, "arxiv_id": "2609.29866", "title": "Beyond Model Size: Redesigning LiSenNet for embedded speech enhancement", "summary": "The method replaces the recurrent bottleneck of a 37k-parameter speech enhancement model with convolutional frequency and temporal mixers, reformulated as static int8-compatible operators with bounded decoder activations, so the model fits a restricted microcontroller NPU. The cost is an architecture-specific redesign and quantization-aware co-design of operator set, quantization ranges, and persistent streaming state, plus a small quality drop after int8 quantization. It fails if the target NPU's operator set differs from the one assumed, if stateless receptive-field recomputation is needed (an order of magnitude slower), or if the target workload is not embedded real-time audio on microcontrollers.", "rationale": "The technique targets microcontroller-class NPUs and static int8 graphs, which are outside the reader's GPU/API production infrastructure and do not transfer to their deployment constraints.", "technique": "NPU-compatible int8 redesign of recurrent speech model", "family": "arquitetura_eficiente", "recommendation": "nao_aplica", "published": "2026-09-24", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2609.29866", "independent_implementations": 0, "total_implementations": 1, "stars": 1, "citations": 0, "claims": [ { "claim_id": "claim-01", "statement": "The method replaces the recurrent bottleneck of a 37k-parameter speech enhancement model with convolutional frequency and temporal mixers, reformulated as static int8-compatible operators with bounded decoder activations, so the model fits a restricted microcontroller NPU. The cost is an architecture-specific redesign and quantization-aware co-design of operator set, quantization ranges, and persistent streaming state, plus a small quality drop after int8 quantization. It fails if the target NPU's operator set differs from the one assumed, if stateless receptive-field recomputation is needed (an order of magnitude slower), or if the target workload is not embedded real-time audio on microcontrollers.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.29866", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "PESQ 3.08 vs 3.01 in FP32 and 3.01 vs 2.93 in int8 relative to the recurrent LiSenNet baseline; real-time factor 0.30 (4.83 ms per 16 ms hop) on STM32N6570-DK.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.29866", "source_page": null, "source_excerpt": "" } ], "exposure_map": [ { "dimension": "quality", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "compute", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "latency", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "operations", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "compatibility", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "security", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "data_and_training", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "reproducibility", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" } ], "risks": [], "minimum_test": [], "open_questions": [], "independent_tests": [], "lab_reproductions": [], "report_available": false, "validation_tier": "unknown", "evidence_tier": "unknown", "infrastructure_basis": "unknown", "equations": [], "equations_status": "not_fetched", "core_kind": "", "equations_fetched_at": "", "equations_source_url": "", "equations_source_sha256": "", "exposures_reviewed_at": "", "exposures_source_sha256": "" }