SpiritLmContent<T>

Models & Types · AiDotNet.TextToSpeech.MultiModal

A piece of SpiritLM’s input or output: text, or 16 kHz speech.

Constructors

Constructor Summary
SpiritLmContent()

Properties

Property Summary
Tensor<T> Audio { get; set; } The speech, when this is speech.
bool IsSpeech { get; } Whether this is speech.
string SpeechTokens { get; set; } The speech tokens ([Hu…], with [Pi…] and [St…] for Expressive) of generated speech.
string Text { get; set; } The text, when this is text.

Methods

Method Summary
SpiritLmContent<T> FromSpeech(Tensor<T> audio) Speech content (16 kHz).
SpiritLmContent<T> FromText(string text) Text content.