Struct KokoroTts 

pub struct KokoroTts {
    jobs: Option<UnboundedSender<Job>>,
    worker: Mutex<Option<JoinHandle<()>>>,
}

A loaded Kokoro-82M pipeline. One render thread per engine, so a second utterance queues behind the first; the thread owns the ONNX session and the runtime the engine’s async API needs.

Kokoro renders a whole clip, so one speak answers one chunk.

Fields

jobs: Option<UnboundedSender<Job>>

worker: Mutex<Option<JoinHandle<()>>>

Implementations

impl KokoroTts

pub fn load(files: &ModelFiles) -> Result<Self, Error>

Load the model and every voice in files.voices, on a thread of their own. A caller may warm it ahead of the first say, and a failed warm is retried by the next one. Every declared digest is checked before the load.

Trait Implementations

impl Drop for KokoroTts

fn drop(&mut self)

Executes the destructor for this type.

fn pin_drop(self: Pin<&mut Self>)

🔬This is a nightly-only experimental API. (pin_ergonomics)

Execute the destructor for this type, but different to Drop::drop, it requires self to be pinned.

impl Engine for KokoroTts

fn speak(&self, request: SpeakRequest) -> Receiver<Result<Chunk, Error>>

Render one utterance. Chunks arrive in order and the channel closes when the render finishes or fails; the host may stop reading at any time.