MoshiConversation<T>

Models & Types · AiDotNet.TextToSpeech.MultiModal

A training example for Moshi’s language model: Moshi’s (the main speaker’s) 24 kHz speech, the user’s if there is one, and the main speaker’s words with their times.

Constructors

Constructor Summary
MoshiConversation()

Properties

Property Summary
IReadOnlyList<MoshiWord> Words { get; set; } The main speaker’s words and their start and end times; null trains the audio with the text stream masked.
Tensor<T> Moshi { get; set; } The main speaker’s speech.
Tensor<T> User { get; set; } The other side’s speech, or null for single-stream audio (pre-training uses the main stream only).