FluidInference / FluidAudio

Swift SDK running open-source speech models (ASR, TTS, VAD, speaker diarization) locally on Apple devices via CoreML and the Neural Engine.