LibriSpeechDataLoader<T>

Models & Types · AiDotNet.Data.Audio.Benchmarks

Loads the LibriSpeech automatic speech recognition dataset (~1000 hours of 16kHz English speech).

Inheritance: DataLoaderBase<T> → InputOutputDataLoaderBase<T, Tensor<T>, Tensor<T>> → LibriSpeechDataLoader<T>

How It Works

LibriSpeech expects:

Features are raw waveform samples as Tensor[N, MaxSamples] (mono, 16kHz). Labels are transcript text encoded as character indices in Tensor[N, MaxTextLen]. For pre-converted WAV files, audio is loaded directly. FLAC requires pre-conversion to WAV.

Constructors

Constructor Summary
LibriSpeechDataLoader(LibriSpeechDataLoaderOptions options) Creates a new LibriSpeech data loader.

Properties

Property Summary
int FeatureCount { get; }
int OutputDimension { get; }
int TotalCount { get; }
string Description { get; }
string Name { get; }

Methods

Method Summary
ValueTuple<IInputOutputDataLoader<T, Tensor<T>, Tensor<T>>, IInputOutputDataLoader<T, Tensor<T>, Tensor<T>>, IInputOutputDataLoader<T, Tensor<T>, Tensor<T>>> Split(double trainRatio, double validationRatio, int? seed)