Module speech 

Speech processing module for Gestura

This module provides core speech-to-text and LLM processing functionality. It is designed to be used by both GUI and CLI applications.

The Tauri-specific event handling (window management, event emission) should be implemented in the gestura-gui crate.

Structs

LlmResponse

Result of LLM processing

SpeechConfig

Speech processing configuration

SpeechProcessor

Core speech processor without Tauri dependencies

TranscriptionResult

Result of speech transcription

Functions

load_audio_samples_16khz_mono

Load audio file and convert to 16kHz mono f32 samples.

resolve_whisper_model_path

Resolve the path to the local Whisper model file.

resolve_whisper_model_path_with_override

Resolve the path to a local Whisper model file, with an optional session-scoped override.