speech_fluidaudio

FluidAudio adapters for the provider-neutral speech_core contracts.

The package exposes streaming and batch STT, VAD, end-of-utterance detection, batch diarization, and TTS while keeping fluidaudio_dart types inside the adapter boundary.

import 'package:speech_fluidaudio/speech_fluidaudio.dart';

final provider = FluidAudioSpeechProvider(runtime: FluidNativeRuntime());

Capture and playback remain in audio_flutter. This lets one capture feed FluidAudio, recording, meters, or another cloud/local provider through the same audio_kit_graph.

FluidAudio inference is on-device and currently targets supported Apple platforms.