//clika-runtime/io.clika.modelverse/TtsModel

TtsModel

[common]
class TtsModel : PreTrainedModel

A text-to-speech model loaded from its files on disk, on the CPU.

Open it with open from a snapshot directory. Hand it a sentence and a reference voice with speak; the clip arrives through the listener on the model's own worker thread, one utterance at a time per model. close frees the weights and the scratch files the model wrote.

Types

Name	Summary
Companion	[common] object Companion

Properties

Name	Summary
capabilities	[common] val capabilities: Capabilities What the model takes and gives, read when it opened: Capabilities.needsReference says whether every utterance needs a VoiceReference, Capabilities.speakKnobs the knobs and ranges speak accepts.
config	[common] open override val config: PretrainedConfig The checkpoint's identity, read when the model opened.

Functions

Name	Summary
close	[common] open fun close()
residency	[common] open override fun residency(): ResidencyReport Free the model and the scratch files it wrote. A running request is canceled first and the call waits for it to return; called from inside a listener callback, the free runs right after that callback's request ends instead. Idempotent.
speak	[common] fun speak(text: String, voice: VoiceReference?, options: SpeakOptions, listener: SpeakListener): SpeakHandle Speak text in the voice voice names, under options. Returns at once; the worker runs the request and calls listener on its thread: the utterance as one SpeechClip through SpeakListener.onClip, then SpeakListener.onDone. A null voice asks for the model's own voice, which a model whose Capabilities.needsReference is true refuses. A knob under its range, a text past the model's window, an utterance the speech window cuts, or a language the model does not interpret ends in SpeakListener.onError with the library's message.