start
fun startAsync(preferredPort: Int, whisperRelease: String, whisperReleaseBaseUrl: String, modelUrl: String, modelsDir: File, binDir: File, whisperDir: File, noGpu: Boolean, threadCount: Int?, maxRAMPercent: Double, maxComputePercent: Double, onReady: (port: Int) -> Unit)
Avvia l'inizializzazione in background: scarica il file binario e il modello del server Whisper, garantisce ffmpeg, avvia il server e avvia il monitor delle risorse.
Parameters
preferred
The TCP port to try first.
whisper
whisper.cpp release tag (e.g. "v1.7.6").
whisper
Base URL prefix for whisper.cpp release downloads.
model
URL of the GGML model to download.
models
Directory to store downloaded models.
bin
Directory to store extracted binaries.
whisper
Root directory for all Whisper artifacts.
no
When true, GPU acceleration is disabled.
thread
Explicit thread count, or null to auto-detect.
max
RAM usage threshold for resource gating.
max
Compute usage threshold for resource gating.
on
Called when the server is fully ready (for updating static port fields).