LjSpeechDataLoader<T>

Models & Types · AiDotNet.Data.Audio.Benchmarks

Loads the LJSpeech 1.1 single-speaker TTS corpus (Ito & Johnson 2017).

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

How It Works

Expects:

Auto-downloads the canonical keithito.com tar.bz2. Features Tensor[N, MaxTextLength] (encoded text tokens); labels Tensor[N, MaxAudioSamples] (audio waveform, mono, 22.05 kHz, [-1, 1]). LJSpeech has no canonical train/val/test splits — this loader applies a deterministic 90/5/5 by row index in the metadata file.

Constructors

Constructor Summary
LjSpeechDataLoader(LjSpeechDataLoaderOptions options)

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)