#FluidAudio
📦 vikas5914/fluidaudio v1.1.0

On-device speech recognition for NativePHP Mobile via FluidAudio (Apple Core ML)

🔗 https://github.com/vikas5914/fluidaudio
September 7, 2026 at 1:35 AM
April 15, 2026 at 9:09 AM
🎉 Countdown to NAMM 2025! Fluid Audio is bringing new studio monitors & pro audio gear to one of the biggest music events.
📍 Booth #15113, ACC North Level 1, Jan 23–25, Anaheim, CA.
Producers, engineers & audio lovers—come experience sound like never before! 👋
#NAMM2025 #FluidAudio #nammshow
January 11, 2025 at 10:48 AM
Dolby Atmos on a budget!

7.1.4: Fluid Audio FX80 + FC10S for $3274.

✨ Why FX80? Precise imaging, wide sweet spot, perfect for Atmos.

Front, Side, Rear, Top = FX80
Sub = FC10S

Pro Tools Ultimate + Avid S1 power this affordable Atmos solution!

#FluidAudio #DolbyAtmos #FX80 #FC10S #ImmersiveAudio
December 6, 2024 at 12:28 PM
Fatbobman’s Swift Weekly #115 is out!
A Sunday Incident

🌠 watchOS Development Pitfalls and Tips
📱 My journey to Swift 6 and Strict Concurrency
📗 TCA Architecture: A Glorified Antipattern
💬 FluidAudio

and more...
weekly.fatbobman.com/p/fatbobman...
Fatbobman's Swift Weekly #115
A Sunday Incident
weekly.fatbobman.com
December 15, 2025 at 12:23 PM
iOS devs, I got a question: what’s the fastest/most reliable way to get realtime transcriptions on iOS 17+?
I’ve gotten mixed success with FluidAudio (parakeet), doesn’t seem like the accuracy is high enough.
iOS 26’s transcription is pretty good, but older devices are out…
December 10, 2025 at 10:20 PM
Next step to my 7.1.4 setup.

#fluidaudio #dolbyatmos #studio
August 12, 2025 at 5:21 AM
Mein kleines Studio verwandelt sich in den nächsten Tagen von dem gängigen Stereo zum 7.1.4 Atmos Studio. Vermutlich das erste in meiner Stadt, vielleicht sogar im Kreis. Es liegt also so einiges an Arbeit an...

#fiedleraudio #dolbyatmos #dolbyatmosmusic #fluidaudio
June 26, 2025 at 9:51 AM
見てる: "FluidInference/FluidAudio: Frontier CoreML audio models in your apps — text-to-speech, speech-to-text, voice activity detection, and speaker diarization. In Swift, powered by SOTA open source." https://github.com/FluidInference/FluidAudio
May 7, 2026 at 1:57 PM
got it. that makes more sense. how's FluidAudio holding up once you get into messy overlaps?
March 23, 2026 at 6:36 PM
how's FluidAudio handling speaker diarization quality? i've found speaker separation is the weakest link in local pipelines. transcription accuracy is good enough now but attributing the right words to the right speaker still breaks down with overlapping speech.
March 20, 2026 at 2:27 PM
FluidAudio is a Swift SDK allowing efficient local audio AI on Apple devices, offering models for speech recognition, text-to-speech, and voice detection, optimized for the Apple Neural Engine, and easy for macOS and iOS integration. https://github.com/FluidInference/FluidAudio
GitHub - FluidInference/FluidAudio: Frontier CoreML audio models in your apps — text-to-speech, speech-to-text, voice activity detection, and speaker diarization. In Swift, powered by SOTA open source. · GitHub
FluidAudio is a Swift SDK allowing efficient local audio AI on Apple devices, offering models for speech recognition, text-to-speech, and voice detection, optimized for the Apple Neural Engine, and ea
github.com
June 12, 2026 at 5:20 AM
The approach is pretty cool:
- Transcribe using an on-device model (I'm using nvidia parakeet v3 via FluidAudio library)
- When a ? is detected, ask the agent. Not perfect, but enough to capture the magic
- When done, send both your transcript and the answers to make a formatted note with Gemini v3
January 12, 2026 at 3:31 AM
ah no, Apple Speech only for the transcription, before I used WhisperKit. For Diarization I use FluidAudio
March 23, 2026 at 7:17 AM
The goal was simple: make it easy to import a recording, get a transcript with speaker separation, and ask questions about it without shipping the content to a hosted LLM.

recall combines:

- WhisperKit for transcription
- FluidAudio for speaker diarization
March 20, 2026 at 10:00 AM
Cool. I use sennheiser open backs and they work well, even for bass.
If I get a larger studio, I'll get a sub! I was eyeing the hs8, but ended up with decent workable monitors, coaxials, from fluidaudio, with a really wide sweet spot. Eartraining for sure, it must be the most overlooked subject!
May 15, 2023 at 6:52 PM
Look at the amount of transcription and meeting software that appeared thanks to tooling like FluidAudio.

Now what? Only few will stand the test of time and even fewer will have the persistence needed to survive.
June 15, 2026 at 7:46 PM
A new AI review! FluidInference/FluidAudio ⭐4.0/5.0
FluidAudio is a well-scoped, production-oriented Swift SDK for fully on-device audio AI on iOS/macOS, focused on **ASR (batch + streaming)**, **TTS**, **VAD**, and **speaker diarization/embeddings**...
https://gitrated.com/FluidInference/FluidAudio
March 22, 2026 at 6:11 AM
FluidInference/FluidAudio: Frontier CoreML audio models in your apps — text-to-speech, speech-to-text, voice activity detection, and speaker diarization. In Swift, powered by SOTA open source.
FluidInference/FluidAudio: Frontier CoreML audio models in your apps — text-to-speech, speech-to-text, voice activity detection, and speaker diarization. In Swift, powered by SOTA open source.
https://github.com/FluidInference/FluidAudio
github.com
June 10, 2026 at 1:10 AM