AudioEnhancementTransformer<T, TInput>
Models & Types · AiDotNet.Preprocessing.Audio
A preprocessing step that runs a configured IAudioEnhancer over audio-tensor inputs,
so audio enhancement composes with the rest of the preprocessing pipeline and is applied consistently to
both training and inference data.
Implements: IDataTransformer<T, TInput, TInput>
For Beginners
This lets you drop a noise-reducer (or dereverb / echo-canceller) into the same preprocessing chain you use for scaling and imputing. Because it is a fitted pipeline step, the enhancer learns its noise profile from your training audio and then cleans every batch — training and live — the same way, instead of being a separate one-off pass that can treat train and test data differently.
Constructors
| Constructor | Summary |
|---|---|
AudioEnhancementTransformer(IAudioEnhancer<T> enhancer) |
Creates an enhancement pipeline step wrapping the given enhancer. |
Properties
| Property | Summary |
|---|---|
bool IsFitted { get; set; } |
|
bool SupportsInverseTransform { get; } |
|
int[] ColumnIndices { get; } |
Methods
| Method | Summary |
|---|---|
TInput FitTransform(TInput data) |
|
TInput InverseTransform(TInput data) |
|
TInput Transform(TInput data) |
|
string[] GetFeatureNamesOut(string[] inputFeatureNames) |
|
void Fit(TInput data) |