Crate denoize 

denoize — pure-Rust audio denoiser built for the world’s highest fidelity.

Goal: transparent, artifact-free restoration that preserves timbre, transients, dynamics, and “air” better than any classical offline tool.

Implemented technologies

Classical DSP (always available)

• STFT/ISTFT + Perfect Reconstruction OLA（高オーバーラップ対応）
• IMCRA/MCRA ノイズ推定 + Spectral Flatness プロファイル + Anchoring
• Ephraim-Malah Decision-Directed SNR
• 8種類のゲイン推定器（OMLSA, LogMMSE, MMSE-STSA, Wiener, SpecSub + 非線形/幾何学的）
• Attack/Release + Cepstral Smoothing + Transient Protection
• 高度窓関数: Kaiser / Flat-top / DPSS
• マルチバンドスペクトルサブトラクション
• 知覚重み付け（Bark帯域）+ 音楽ノイズ抑制ポストフィルタ

Input / output codecs (built-in, no ffmpeg)

• Decode: WAV / MP3 (symphonia, including Xing/LAME gapless timing, with a bounded raw-stream compatibility fallback) / M4A (Pure-Rust AAC-LC and ALAC with v0/v1 unity-rate edit-list presentation timing)
• Encode: WAV / MP3 (shine-rs) / M4A (oxideav-aac Pure-Rust AAC-LC)
• Decoded to f64 PCM at native sample rate (no extra quantisation)

Optional AI backends (feature-gated)

• rnnoise feature: RNNoise via nnnoiseless (pure-Rust)
• deepfilter feature: DeepFilterNet v3 via tract ONNX
• onnx feature: user-supplied waveform ONNX models via tract (including a reusable, contract-checked loaded-model API)
• mpsenet feature: MP-SENet compressed-magnitude/phase ONNX adapter
• bsrnn feature: ESPnet BSRNN spectral ONNX adapter
• mossformer2 feature: ClearerVoice MossFormer2 48 kHz ONNX adapter
• sgmse feature: SGMSE+ iterative diffusion ONNX adapter

Build with all backends: cargo build --release --features full

Re-exports

pub use atomic_output::AtomicOutput;

pub use atomic_output::CommitMode;

pub use audio::ensure_memory_limit;

pub use audio::estimate_audio_memory_bytes;

pub use audio::estimate_audio_working_set_bytes;

pub use audio::estimate_file_memory_bytes;

pub use audio::estimate_session_memory_bytes;

pub use audio::estimate_stream_memory_bytes;

pub use audio::estimate_stream_memory_bytes_checked;

pub use audio::inspect_wav_session;

pub use audio::read_audio;

pub use audio::read_audio_from_session;

pub use audio::read_audio_from_session_with_limits;

pub use audio::read_audio_with_limits;

pub use audio::read_audio_with_metadata_limits;

pub use audio::read_wav;

pub use audio::read_wav_bytes;

pub use audio::read_wav_bytes_with_limits;

pub use audio::read_wav_from_session;

pub use audio::read_wav_from_session_with_limits;

pub use audio::read_wav_with_limits;

pub use audio::sanitize_sample;

pub use audio::write_audio;

pub use audio::write_wav;

pub use audio::write_wav_bytes;

pub use audio::write_wav_channel_mask;

pub use audio::Audio;

pub use audio::WavStreamInfo;

pub use audio::WavStreamReader;

pub use audio::WavStreamWriter;

pub use backend::gtcrn::GtcrnModel;

pub use backend::gtcrn::GtcrnStream;

pub use backend::onnx::OnnxWaveformContract;

pub use backend::onnx::OnnxWaveformLayout;

pub use backend::onnx::OnnxWaveformModel;

pub use backend::decode_mid_side;

pub use backend::encode_mid_side;

pub use backend::Backend;

pub use backend::BackendOptions;

pub use backend::BackendSession;

pub use backend::ChannelMode;

pub use backend::OnnxModelConfig;

pub use backend::SgmseProfile;

pub use backend::StreamingBackendSession;

pub use benchmark::ArtifactReport;

pub use benchmark::BenchmarkReport;

pub use benchmark::ComparisonReport;

pub use channel_layout::ChannelLayout;

pub use channel_layout::ChannelMask;

pub use channel_layout::ChannelPosition;

pub use channel_layout::PanInfo;

pub use config::ConfigError;

pub use config::ResourcePlan;

pub use daw::read_daw_preset;

pub use daw::read_daw_session;

pub use daw::write_daw_preset;

pub use daw::write_daw_session;

pub use daw::DawParameters;

pub use daw::DawPortConfiguration;

pub use daw::DawPreset;

pub use daw::DawRealtimeParameters;

pub use daw::DawRealtimeProcessor;

pub use daw::DawSessionState;

pub use daw::DAW_FIXED_LATENCY_MILLIS;

pub use daw::DAW_LATENCY_POLICY;

pub use daw::DAW_MAX_SAMPLE_RATE;

pub use daw::DAW_PLUGIN_ID;

pub use daw::DAW_PRESET_SCHEMA;

pub use daw::DAW_PRESET_SCHEMA_VERSION;

pub use daw::DAW_SESSION_SCHEMA;

pub use daw::DAW_SESSION_SCHEMA_VERSION;

pub use decode::decode_file;

pub use decode::decode_file_from_session_with_limits;

pub use decode::decode_file_with_limits;

pub use decode::decode_file_with_metadata_limits;

pub use decode::inspect_audio_stream_session;

pub use decode::probe_file;

pub use decode::probe_file_from_session_with_limits;

pub use decode::probe_file_with_limits;

pub use decode::probe_file_with_metadata_limits;

pub use decode::AudioCodec;

pub use decode::AudioFormat;

pub use decode::AudioProbe;

pub use decode::AudioStreamInfo;

pub use decode::AudioStreamReader;

pub use decode::DecodeLimits;

pub use decode::DecodedPcm;

pub use denoiser::Denoiser;

pub use denoiser::DenoiserConfig;

pub use denoiser::Preset;

pub use denoiser::ProcessingMode;

pub use denoiser::StreamingDenoiser;

pub use diagnostics::assess_file_with_options;

pub use diagnostics::compare_files_with_options;

pub use diagnostics::diagnose_audio;

pub use diagnostics::diagnose_file;

pub use diagnostics::diagnose_file_with_options;

pub use diagnostics::AssessmentComparison;

pub use diagnostics::AssessmentReport;

pub use diagnostics::DiagnosticFinding;

pub use diagnostics::DiagnosticInput;

pub use diagnostics::DiagnosticMetrics;

pub use diagnostics::DiagnosticOptions;

pub use diagnostics::DiagnosticReport;

pub use diagnostics::NoReferenceQuality;

pub use diagnostics::ASSESSMENT_SCHEMA;

pub use diagnostics::ASSESSMENT_SCHEMA_VERSION;

pub use diagnostics::DIAGNOSTIC_SCHEMA;

pub use diagnostics::DIAGNOSTIC_SCHEMA_VERSION;

pub use encode::estimate_spooled_stream_output_bytes;

pub use encode::estimate_stream_encode_additional_bytes;

pub use encode::estimate_stream_encode_output_bytes;

pub use encode::estimate_stream_encode_temporary_bytes;

pub use encode::estimate_stream_output_verification_bytes;

pub use encode::verify_stream_output_file;

pub use encode::AacEncoder;

pub use encode::AudioStreamWriter;

pub use encode::DownmixMode;

pub use encode::EncodeOptions;

pub use encode::OutputFormat;

pub use encode::SpooledAudioStreamWriter;

pub use encode::StreamEncodeLimits;

pub use encode::StreamEncodeSpec;

pub use encode::StreamOutputVerification;

pub use encode::StreamPcmSpool;

pub use evaluation::compare_evaluation_results;

pub use evaluation::run_evaluation;

pub use evaluation::validate_evaluation_corpus;

pub use evaluation::verify_evaluation_result;

pub use evaluation::write_signed_evaluation_result;

pub use evaluation::EvaluationArtifact;

pub use evaluation::EvaluationCase;

pub use evaluation::EvaluationCaseResult;

pub use evaluation::EvaluationComparisonReport;

pub use evaluation::EvaluationCorpusValidation;

pub use evaluation::EvaluationEnvironment;

pub use evaluation::EvaluationLicense;

pub use evaluation::EvaluationManifest;

pub use evaluation::EvaluationPolicy;

pub use evaluation::EvaluationRecipe;

pub use evaluation::EvaluationResultPayload;

pub use evaluation::EvaluationSource;

pub use evaluation::EvaluationThreshold;

pub use evaluation::EvaluationVerificationReport;

pub use evaluation::ListeningEvidence;

pub use evaluation::ListeningPolicy;

pub use evaluation::ListeningProtocol;

pub use evaluation::ListeningTestResult;

pub use evaluation::ObjectiveEvaluationMetrics;

pub use evaluation::OutputQualityMetrics;

pub use evaluation::PerformanceMetrics;

pub use evaluation::RegressionDirection;

pub use evaluation::RegressionOutcome;

pub use evaluation::RegressionTolerance;

pub use evaluation::SignalPreparation;

pub use evaluation::SignedEvaluationResult;

pub use evaluation::ThresholdAggregation;

pub use evaluation::ThresholdOperator;

pub use evaluation::ThresholdOutcome;

pub use evaluation::EVALUATION_COMPARISON_SCHEMA;

pub use evaluation::EVALUATION_CORPUS_SCHEMA;

pub use evaluation::EVALUATION_CORPUS_VERIFICATION_SCHEMA;

pub use evaluation::EVALUATION_RESULT_SCHEMA;

pub use evaluation::EVALUATION_SCHEMA_VERSION;

pub use evaluation::EVALUATION_VERIFICATION_SCHEMA;

pub use evaluation::LISTENING_RESULT_SCHEMA;

pub use execution::execution_item_id;

pub use execution::export_receipt_public_key;

pub use execution::generate_receipt_keypair;

pub use execution::portable_file_locator;

pub use execution::portable_locator;

pub use execution::write_execution_plan;

pub use execution::write_new_receipt_keypair;

pub use execution::write_receipt_trust_policy;

pub use execution::write_signed_receipt;

pub use execution::ExecutionKind;

pub use execution::ExecutionPlan;

pub use execution::ExecutionPlanItem;

pub use execution::ExecutionReceiptPayload;

pub use execution::PlannedArtifact;

pub use execution::PlannedOutput;

pub use execution::PlannedResources;

pub use execution::ReceiptItem;

pub use execution::ReceiptOutput;

pub use execution::ReceiptPublicKey;

pub use execution::ReceiptSecretKey;

pub use execution::ReceiptSignature;

pub use execution::ReceiptTrustPolicy;

pub use execution::ReceiptVerificationReport;

pub use execution::SignedExecutionReceipt;

pub use execution::VerifiedReceiptItem;

pub use execution::EXECUTION_PLAN_SCHEMA;

pub use execution::EXECUTION_RECEIPT_SCHEMA;

pub use execution::EXECUTION_SCHEMA_VERSION;

pub use execution::RECEIPT_PUBLIC_KEY_SCHEMA;

pub use execution::RECEIPT_SECRET_KEY_SCHEMA;

pub use execution::RECEIPT_TRUST_POLICY_SCHEMA;

pub use execution::RECEIPT_VERIFICATION_SCHEMA;

pub use execution::STREAM_EXECUTION_PLAN_SCHEMA;

pub use execution::STREAM_EXECUTION_RECEIPT_SCHEMA;

pub use execution::STREAM_EXECUTION_SCHEMA_VERSION;

pub use execution::STREAM_RECEIPT_VERIFICATION_SCHEMA;

pub use gain::Algorithm;

pub use gain::SpecSubLaw;

pub use hardware::backend_supports_acceleration;

pub use hardware::hardware_capabilities;

pub use hardware::select_accelerator;

pub use hardware::select_accelerator_for_options;

pub use hardware::AcceleratorFallback;

pub use hardware::AcceleratorPreference;

pub use hardware::AcceleratorRuntime;

pub use hardware::AcceleratorSelection;

pub use hardware::BackendCapability;

pub use hardware::HardwareCapabilities;

pub use hardware::RuntimeCapability;

pub use hardware::HARDWARE_SCHEMA;

pub use hardware::HARDWARE_SCHEMA_VERSION;

pub use input::AudioInputSession;

pub use input::StreamSpoolLimits;

pub use model_package::build_runtime_model_package;

pub use model_package::build_runtime_model_package_v2;

pub use model_package::inspect_runtime_model_package;

pub use model_package::RuntimeModelAxisContractV2;

pub use model_package::RuntimeModelChannelContractV2;

pub use model_package::RuntimeModelChannelRoleContractV2;

pub use model_package::RuntimeModelComponentContractV2;

pub use model_package::RuntimeModelFileContract;

pub use model_package::RuntimeModelFrontendContract;

pub use model_package::RuntimeModelFrontendContractV2;

pub use model_package::RuntimeModelGeometryContractV2;

pub use model_package::RuntimeModelLatencyContractV2;

pub use model_package::RuntimeModelLicenseContract;

pub use model_package::RuntimeModelLicenseContractV2;

pub use model_package::RuntimeModelMicrophonePositionV2;

pub use model_package::RuntimeModelNumericalCaseV1;

pub use model_package::RuntimeModelNumericalTensorV1;

pub use model_package::RuntimeModelNumericalToleranceV1;

pub use model_package::RuntimeModelNumericalVectorsV1;

pub use model_package::RuntimeModelPackage;

pub use model_package::RuntimeModelPackageInfo;

pub use model_package::RuntimeModelPackageManifest;

pub use model_package::RuntimeModelPackageManifestV2;

pub use model_package::RuntimeModelPackageReader;

pub use model_package::RuntimeModelPackageV2Info;

pub use model_package::RuntimeModelPrecisionProfileContractV2;

pub use model_package::RuntimeModelProvenanceContractV2;

pub use model_package::RuntimeModelResourceContract;

pub use model_package::RuntimeModelRuntimeContract;

pub use model_package::RuntimeModelRuntimeContractV2;

pub use model_package::RuntimeModelStatePairContractV2;

pub use model_package::RuntimeModelTensorContract;

pub use model_package::RuntimeModelTensorContractV2;

pub use model_package::RuntimeModelTensorSetContractV2;

pub use model_package::RuntimeModelTrainingDatasetContractV2;

pub use model_package::RUNTIME_MODEL_PACKAGE_SCHEMA;

pub use model_package::RUNTIME_MODEL_PACKAGE_SCHEMA_V2;

pub use model_package::RUNTIME_MODEL_PACKAGE_VERSION;

pub use model_package::RUNTIME_MODEL_PACKAGE_VERSION_V2;

pub use neural_daw::neural_daw_chunk_frames;

pub use neural_daw::neural_daw_latency_frames;

pub use neural_daw::neural_daw_latency_millis;

pub use neural_daw::read_neural_daw_session;

pub use neural_daw::write_neural_daw_session;

pub use neural_daw::NeuralDawOverloadFallback;

pub use neural_daw::NeuralDawParameters;

pub use neural_daw::NeuralDawPortConfiguration;

pub use neural_daw::NeuralDawSessionState;

pub use neural_daw::NEURAL_DAW_CHUNK_MILLIS;

pub use neural_daw::NEURAL_DAW_LATENCY_CHUNKS;

pub use neural_daw::NEURAL_DAW_LATENCY_POLICY;

pub use neural_daw::NEURAL_DAW_MAX_SAMPLE_RATE;

pub use neural_daw::NEURAL_DAW_MODEL_ID;

pub use neural_daw::NEURAL_DAW_MODEL_SHA256;

pub use neural_daw::NEURAL_DAW_PLUGIN_ID;

pub use neural_daw::NEURAL_DAW_SESSION_SCHEMA;

pub use neural_daw::NEURAL_DAW_SESSION_SCHEMA_VERSION;

pub use project::assemble_project_timeline;

pub use project::build_project_bundle;

pub use project::import_project_bundle;

pub use project::inspect_project_bundle;

pub use project::inspect_project_source;

pub use project::project_artifact_reference;

pub use project::relocate_project_source;

pub use project::run_project_batch;

pub use project::validate_project_files;

pub use project::write_project_execution_plan;

pub use project::write_project_manifest;

pub use project::write_signed_project_execution_receipt;

pub use project::ProjectArtifactReference;

pub use project::ProjectBatchItemReport;

pub use project::ProjectBatchReport;

pub use project::ProjectBatchRequest;

pub use project::ProjectBundleBinding;

pub use project::ProjectBundleBindingKind;

pub use project::ProjectBundleBuildOptions;

pub use project::ProjectBundleFileInfo;

pub use project::ProjectBundleImportReport;

pub use project::ProjectBundleInfo;

pub use project::ProjectExecutionPlan;

pub use project::ProjectExecutionReceiptPayload;

pub use project::ProjectManifest;

pub use project::ProjectModelReference;

pub use project::ProjectReceiptVerificationReport;

pub use project::ProjectRenderReport;

pub use project::ProjectSelection;

pub use project::ProjectSource;

pub use project::ProjectSourceInspection;

pub use project::ProjectTimeline;

pub use project::ProjectValidationReport;

pub use project::SignedProjectExecutionReceipt;

pub use project::PROJECT_BATCH_SCHEMA;

pub use project::PROJECT_BUNDLE_IMPORT_SCHEMA;

pub use project::PROJECT_BUNDLE_SCHEMA;

pub use project::PROJECT_EXECUTION_PLAN_SCHEMA;

pub use project::PROJECT_EXECUTION_RECEIPT_SCHEMA;

pub use project::PROJECT_MANIFEST_SCHEMA;

pub use project::PROJECT_MANIFEST_SCHEMA_VERSION;

pub use project::PROJECT_RECEIPT_VERIFICATION_SCHEMA;

pub use project::PROJECT_RENDER_SCHEMA;

pub use project::PROJECT_VALIDATION_SCHEMA;

pub use project::PROJECT_WATCH_CYCLE_SCHEMA;

pub use quality::QualityMetrics;

pub use recommendation::recommend_audio;

pub use recommendation::recommend_file;

pub use recommendation::recommend_file_with_options;

pub use recommendation::run_device_calibration;

pub use recommendation::CalibrationEvidence;

pub use recommendation::RecommendationCandidate;

pub use recommendation::RecommendationDecision;

pub use recommendation::RecommendationDevice;

pub use recommendation::RecommendationGoal;

pub use recommendation::RecommendationInput;

pub use recommendation::RecommendationMaterial;

pub use recommendation::RecommendationOptions;

pub use recommendation::RecommendationReason;

pub use recommendation::RecommendationReport;

pub use recommendation::RECOMMENDATION_SCHEMA;

pub use recommendation::RECOMMENDATION_SCHEMA_VERSION;

pub use region::PresentationRegion;

pub use region::PRESENTATION_REGION_SCHEMA;

pub use region::PRESENTATION_REGION_SCHEMA_VERSION;

pub use resource::estimate_backend_session_request;

pub use resource::estimate_backend_worker_gpu_memory_bytes;

pub use resource::estimate_backend_worker_memory_bytes;

pub use resource::estimate_gpu_session_bytes;

pub use resource::estimate_gpu_worker_bytes;

pub use resource::estimate_model_session_bytes;

pub use resource::estimate_temporary_bytes;

pub use resource::metadata_limits_after_retained_memory;

pub use resource::metadata_limits_for_available_memory;

pub use resource::ResourceGovernor;

pub use resource::ResourceLimits;

pub use resource::ResourcePermit;

pub use resource::ResourceRequest;

pub use resource::ResourceUsage;

pub use restoration::estimate_restoration_memory_bytes;

pub use restoration::restore_audio;

pub use restoration::DeclickConfig;

pub use restoration::DeclipConfig;

pub use restoration::DehumConfig;

pub use restoration::RestorationConfig;

pub use restoration::RestorationMask;

pub use restoration::RestorationMaskRun;

pub use restoration::RestorationMaskState;

pub use restoration::RestorationMode;

pub use restoration::RestorationOperation;

pub use restoration::RestorationOperationDetails;

pub use restoration::RestorationOperationReport;

pub use restoration::RestorationReport;

pub use restoration::RestorationResult;

pub use restoration::RestorationStatus;

pub use restoration::WindPlosiveConfig;

pub use restoration::WpeChannelMode;

pub use restoration::WpeConfig;

pub use restoration::MAX_RESTORATION_CHANNELS;

pub use restoration::MAX_RESTORATION_MASK_RUNS;

pub use restoration::MAX_RESTORATION_OPERATIONS;

pub use restoration::RESTORATION_MASK_SCHEMA;

pub use restoration::RESTORATION_REPORT_SCHEMA;

pub use restoration::RESTORATION_SCHEMA_VERSION;

pub use target_speaker::TargetSpeakerSession;

pub use target_speaker::estimate_target_speaker_memory_bytes;

pub use target_speaker::sign_target_speaker_promotion_evidence;

pub use target_speaker::SignedTargetSpeakerPromotionEvidence;

pub use target_speaker::TargetSpeakerDecision;

pub use target_speaker::TargetSpeakerEnrollmentSummary;

pub use target_speaker::TargetSpeakerEvidenceIdentity;

pub use target_speaker::TargetSpeakerExtractionConfig;

pub use target_speaker::TargetSpeakerExtractionReport;

pub use target_speaker::TargetSpeakerExtractionResult;

pub use target_speaker::TargetSpeakerMetricOperator;

pub use target_speaker::TargetSpeakerMetricOutcome;

pub use target_speaker::TargetSpeakerModelIdentity;

pub use target_speaker::TargetSpeakerPresence;

pub use target_speaker::TargetSpeakerPresenceAssessment;

pub use target_speaker::TargetSpeakerPromotionEvidencePayload;

pub use target_speaker::TargetSpeakerSafetyGate;

pub use target_speaker::TargetSpeakerSafetyGateKind;

pub use target_speaker::TargetSpeakerSafetyMeasurements;

pub use target_speaker::TargetSpeakerStratumEvidence;

pub use target_speaker::TargetSpeakerStratumKind;

pub use target_speaker::MAX_TARGET_SPEAKER_ENROLLMENT_MILLIS;

pub use target_speaker::MAX_TARGET_SPEAKER_EVIDENCE_METRICS;

pub use target_speaker::MAX_TARGET_SPEAKER_EVIDENCE_STRATA;

pub use target_speaker::MAX_TARGET_SPEAKER_MIXTURE_SECONDS;

pub use target_speaker::MIN_TARGET_SPEAKER_ENROLLMENT_MILLIS;

pub use target_speaker::TARGET_SPEAKER_PROMOTION_EVIDENCE_SCHEMA;

pub use target_speaker::TARGET_SPEAKER_REPORT_SCHEMA;

pub use target_speaker::TARGET_SPEAKER_SCHEMA_VERSION;

pub use universal_restoration::restore_universal_audio;

pub use universal_restoration::estimate_universal_restoration_memory_bytes;

pub use universal_restoration::sign_universal_promotion_evidence;

pub use universal_restoration::SignedUniversalPromotionEvidence;

pub use universal_restoration::UniversalDegradation;

pub use universal_restoration::UniversalDegradationEvidence;

pub use universal_restoration::UniversalMaskRun;

pub use universal_restoration::UniversalMaskState;

pub use universal_restoration::UniversalMetricOperator;

pub use universal_restoration::UniversalMetricOutcome;

pub use universal_restoration::UniversalModelFamily;

pub use universal_restoration::UniversalModelIdentity;

pub use universal_restoration::UniversalPromotionEvidencePayload;

pub use universal_restoration::UniversalRenderRole;

pub use universal_restoration::UniversalRestorationConfig;

pub use universal_restoration::UniversalRestorationDecision;

pub use universal_restoration::UniversalRestorationMask;

pub use universal_restoration::UniversalRestorationReport;

pub use universal_restoration::UniversalRestorationResult;

pub use universal_restoration::UniversalSafetyGate;

pub use universal_restoration::UniversalSafetyGateKind;

pub use universal_restoration::UniversalSafetyMeasurements;

pub use universal_restoration::UniversalStratumEvidence;

pub use universal_restoration::MAX_UNIVERSAL_EVIDENCE_METRICS;

pub use universal_restoration::MAX_UNIVERSAL_EVIDENCE_STRATA;

pub use universal_restoration::MAX_UNIVERSAL_MASK_RUNS;

pub use universal_restoration::UNIVERSAL_PROMOTION_EVIDENCE_SCHEMA;

pub use universal_restoration::UNIVERSAL_RESTORATION_MASK_SCHEMA;

pub use universal_restoration::UNIVERSAL_RESTORATION_REPORT_SCHEMA;

pub use universal_restoration::UNIVERSAL_RESTORATION_SCHEMA_VERSION;

pub use watch::WatchCycleReport;

pub use watch::WatchFolder;

pub use watch::WatchFolderConfig;

pub use watch::WatchFolderJob;

pub use watch::WatchProcessError;

pub use watch::WATCH_CYCLE_SCHEMA;

pub use watch::WATCH_QUARANTINE_SCHEMA;

pub use watch::WATCH_SCHEMA_VERSION;

pub use watch::WATCH_STATE_SCHEMA;

pub use window::WindowParams;

pub use window::WindowType;

Modules

atomic_output

Atomic creation and replacement of filesystem outputs.

audio

Audio file I/O: WAV read/write (hound) + unified decode for MP3/M4A/WAV.

automation

Stable, read-only automation snapshots.

backend

Optional AI denoising backends (feature-gated).

benchmark

Objective quality and stereo-imaging reports.

bessel

Special functions needed by the Ephraim-Malah family of gain estimators: modified Bessel functions of the first kind, orders 0 and 1, and the exponential integral E1.

channel_layout

Standard channel layouts used by the audio pipeline.

config

Shared validation and resource-planning primitives for external input.

daw

Real-time-safe DSP and portable state contracts for the DAW plug-in.

decode

denoize 自作デコード層 — lossless / lossy audio containers を高品質 PCM (f64) へ。

denoiser

The de-noizer: ties together STFT, IMCRA noise estimation, the decision-directed a-priori SNR estimator, the selected spectral gain estimator, attack/release + cepstral gain smoothing, transient protection, and optional pre-emphasis.

diagnostics

Bounded, network-free audio degradation diagnosis and no-reference quality assessment.

encode

Audio encode layer for bounded WAV, FLAC, Ogg Opus, MP3, M4A, and ADTS AAC output.

evaluation

Reproducible licensed-corpus evaluation and signed release evidence.

execution

Stable execution plans, signed receipts, and offline result verification.

fft

A self-contained, allocation-light, iterative radix-2 Cooley-Tukey FFT.

gain

Spectral gain estimators.

hardware

Compute capability discovery and explicit accelerator selection.

input

Stable, regular-file-backed audio input sessions.

ipc

Local authenticated IPC and durable job control.

loudness

EBU R128 / ITU-R BS.1770 loudness measurement and normalization.

metadata

Cross-container audio metadata preservation.

model_package

Signed, self-describing runtime packages for custom waveform models.

models

Signed external-model catalog, verified local cache, and installation provenance.

neural_daw

Portable contracts shared by the off-callback neural DAW plug-in and its CLI/Desktop control surfaces.

noise

Noise-power-spectral-density estimation.

perceptual

Perceptual weighting for spectral gain control.

postfilter

Musical-noise suppression post-filter.

project

Portable, source-bound partial-file timelines.

quality

Full-reference perceptual quality metrics.

recommendation

Network-free, explainable processing recommendations.

region

Stable, source-bound presentation regions.

resample

Band-limited, channel-synchronous sample-rate conversion.

resource

Process-wide admission control for bounded denoising work.

restoration

Deterministic, inspectable audio restoration.

service

Shared application service used by the CLI and graphical frontends.

stft

Short-time Fourier transform engine with perfect-reconstruction overlap-add (OLA) synthesis.

stream

Stateful block-processing API for realtime and pipe integrations.

target_speaker

Fail-closed offline target-speaker extraction and promotion evidence.

universal_restoration

Fail-closed orchestration and evidence contracts for universal speech restoration.

update

Authenticated, recoverable application-update transactions.

vad

Lightweight energy VAD and speech-region segmentation.

watch

Durable local watch-folder automation.

window

Window functions for short-time Fourier analysis/synthesis.

Functions

denoise_audio_with_backend_config

Process already-decoded audio in place. This is the path used by stdin and embedders that do not have filesystem-backed input.

denoise_audio_with_backend_session

Process decoded audio with a prepared backend graph.

denoise_file

Denoise a WAV file end-to-end, writing the result to output.

denoise_file_with_backend

Denoise with an explicit backend (classical / rnnoise / deepfilter).

denoise_file_with_backend_config

Denoise with explicit backend, encoder, and backend-specific model options.

denoise_file_with_backend_opts

Denoise with explicit backend and output encode options.

write_audio_transactional

Encode audio and optional metadata into a staged file, then publish it in one filesystem commit.

write_audio_transactional_as

Encode audio using a format selected during preflight, then publish it in one filesystem commit without re-inferring the codec from the path.