XTTSv2Clone<T>

Models & Types · AiDotNet.TextToSpeech.VoiceCloning

Voice-cloning facade over the canonical XTTS-v2 zero-shot architecture.

Inheritance: NeuralNetworkBase<T> → TtsModelBase<T> → XTTSv2<T> → XTTSv2Clone<T>

Implements: IVoiceCloner<T>

How It Works

XTTS-v2 voice cloning is not a second model. This facade inherits the exact dVAE tokenizer, conditioning encoder/Perceiver, GPT-2, H/ASP speaker encoder and speaker-conditioned HiFi-GAN from XTTSv2 and adapts its reference-audio overload to IVoiceCloner.

Constructors

Constructor Summary
XTTSv2Clone(NeuralNetworkArchitecture<T> architecture, XTTSv2CloneOptions options, IGradientBasedOptimizer<T, Tensor<T>, Tensor<T>> optimizer)
XTTSv2Clone(NeuralNetworkArchitecture<T> architecture, string modelPath, XTTSv2CloneOptions options)

Properties

Property Summary
double MinReferenceDuration { get; }
int SpeakerEmbeddingDim { get; }

Methods

Method Summary
ModelMetadata<T> GetModelMetadata()
Tensor<T> SynthesizeWithVoice(string text, Tensor<T> referenceAudio)