ALTIC

FluidVoice 1.5.12 is out with new models! First local dictation app to ship @cohere Transcribe. Accuracy maxxed and fully open source as-usual. Go check it out and let me know what you think. Github below 👇

@mark_k @OpenAI Gemini 3 flash… sure

@madebycharlie_ @WisprFlow I am building models to get close to bridging the difference. I let it out and there goes their value prop.. thanks for using it and supporting OSS, Charlie

@ALTIC_DEV @WisprFlow I love the app dude. Open Source is the way to move humanity forward and away of tech monopolies profiting from our data. Thanks for building Fluid Voice

Stop giving your personal data to predatory companies like @WisprFlow and instead use free, private and open source alternatives like @ALTIC_DEV - Fluid Voice. Thank me later.

@madebycharlie_ @WisprFlow github.com/altic-dev/Flui… If anyone is interested in not giving up your data and make use of your mac ;)

@luxus @cohere Double press is new hmm.. never cared much as there wasnt much request on that.

its actually one of the better programs.. i felt into the rabithole and tried a lot of these apps :D first thing i miss is the better keybinding behavior in some apps.. i love the fn hold for push to talk and double fn for toggle .. but nice to see that you listen to requests, the additional key for dictate with llm is great

FluidVoice 1.5.12 is out with new models! First local dictation app to ship @cohere Transcribe. Accuracy maxxed and fully open source as-usual. Go check it out and let me know what you think. Github below 👇

@luxus @cohere Did you try the update ;) if you were the one to request this change,. It’s fixed and your wish is granted 👀

@alexabelonix @cohere Thank youuuu :)))

@ALTIC_DEV @cohere love the energy.

At 11pm, I decided to add captions to my video. By 12am, I had built an entire video editing app that generates word-by-word captions from audio — with in-video caption editing too. All with @OpenAI Codex. So… is software engineering dead? Video attached. The wild part: caption editing like this is usually a paid feature in most video apps. I built it for free in an hour.

@cohere Transcribe - one of the most accurate transcription models we have tested till now - has been added to FluidVoice for fast, accurate dictation❤️ And man-oh-man is it incredible🤩 We stress tested it and we loved it. Hope you all love it too! Built for the community, by the community -> FluidVoice

Is anyone interested in @cohere Transcribe?? Heard it’s dope.. i can add it to FluidVoice 👀👀

@AravSrinivas 123

🚨 Mac owners, you NEED to try this: I’ve been using #FluidVoice every single day and it has genuinely changed how I work: I now talk to my Mac way more than I type. It’s the fastest completely free & #openSource offline dictation app #VoiceToText for macOS: ✅ 100% local AI (zero cloud, maximum privacy) ✅ Insanely low latency + beautiful live preview ✅ Global hotkey, types directly into ANY app ✅ Write Mode + powerful Command Mode ✅ Multiple top models (Parakeet, Whisper...) No subscription. No data collection. No compromise. If you use #MacOS , do yourself a favor and try it today: altic.dev/fluid GitHub: github.com/altic-dev/Flui… test it for a few days, then come back and tell me honestly: What do you think? 🔥

@deepfates github.com/altic-dev/Flui… Not perfect but getting there :) a fully free macOS FOSS alternative.

agree, in fact there's tons of software you can use already instead of WisprFlow! I use spokenly and the Nvidia parakeet model which can run on like pretty much any laptop I think and it's better faster local and free. unrelated how do I remove a picture from this tweet

As a single person working on FluidVoice, it’s hard to maintain a flawless product but trying to fix as much as I could 🥲 thanks for trying. It’s insanely hard to make it work well across all devices when it works so perfectly in my local testing. Hope to make it better soon and also working on a local FT model as well for better enhancement! No one should pay for local dictation.

Is Spokenly actually a better UX overall than WisprFlow? I’m probably 99th percentile in voice to text usage and I tried a FOSS tool FluidVoice which lets you use Parakeet locally Yes the model is great, but the product is buggy. FluidVoice has lots of functionality built around the model and the iPhone functionality is solid Stats from 2 weeks of usage 😂