Shypd.aiThe largest open index of AI toolsFluidAudio is a Content & Design tool that provides a Swift SDK for local, low-latency audio AI on Apple devices. It offers text-to-speech, speech-to-text, voice activity detection, and speaker diarization, powered by open-source CoreML models.
Trending
About
FluidAudio is a Swift SDK designed for fully local, low-latency audio AI inference on Apple devices, leveraging the Apple Neural Engine (ANE) for optimal performance and minimal power consumption. This SDK includes state-of-the-art capabilities such as automatic speech recognition (ASR) with models like Parakeet TDT v3 and Parakeet EOU, supporting multiple languages and streaming. It also features inverse text normalization (ITN) for post-processing ASR output, text-to-speech (TTS) with Kokoro and PocketTTS for various languages and voice cloning, and robust speaker diarization for identifying and separating speakers in audio streams. Additionally, it provides voice activity detection (VAD) and speaker embedding extraction. All models are open-source, optimized for ANE, and available on HuggingFace, making FluidAudio ideal for developers building ambient computing and always-on audio applications for iOS and macOS.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending
