I've dictated almost everything for 6 months with Wispr Flow. 44,414 words, 161 wpm, top 0.1% of users.
Last week I tried FluidVoice. Open source, runs local on my Mac, corrects as I speak with no API key, and handles slang better than I expected.
Cancelled my paid plan. If you're on a Mac, this one's for you: altic.dev/fluid
Post
Conversation
Post your reply
This genuinely made my day :,)) Posts like this make last 4 months of sleepless nights, and research to give the best output worth it.
Is it really comparable though? Have tired a few others and whisperflow just had fewer rough edges and was worth the few dollars a month.
Consider me twice bitten once shy.
Founder is very active responding to feedback and bugs. This is core infra now.
Ha, you've got me beat. I use it more for prompts and emails than long form, so my numbers stay modest. Either way, good to see FluidVoice getting a proper workout.
44k words? I do that in 3 days. I have 1M words dictated. 44k? Are you only using it for shorthand texts?
I use it more for prompts and emails than long form, so my numbers stay modest.
Interesting switch after those stats. Local execution and real-time corrections on Mac sound solid technically, but curious how the slang handling holds up on longer sessions versus Wispr Flow's consistency.
I used for longer version for the last one week and it worked well. I am not native English speaker
More From This Author
insforge.dev by is really well done.
Tested it with my coding agent - real backend (db, auth, storage, AI gateway via MCP), shipped a client POC end-to-end. Even sent a small PR to the repo.
Genuinely good work. Loved it.
html-docs.com by is really well done.
I convert my client docs to clean shareable links now instead of sending HTML files. Clients just click and read.
Loved it.
How would you compare this to handy? I’ve been using this 6 months and mega happy:
I switched to Fluid too last week. Ran out of patience with WisprFlow lag (even with short phrases sometimes) or total outages.
Is there a way to ensure it doesn't redownload any existing and already downloaded local models and is pointed towards using your already downloaded models to avoid using double the disk space? What's your recommended way to set that up so a user can conserve disk space?
Default macOS speech to text works really well for me. Map it fn key and it is super straightforward. No third party needed
Will this eventually work on iPhone and iPad? It would be great to have this for the iPad, maybe better than for my Mac.
Those are rookie numbers. This is ~13 months on .
If I only used on my Mac I’d consider open source/local. But $15 for polish + “it just works” + iOS app is an easy choice.
This isn’t just an Exclusive remaster. It’s a full remake of Under a Killing Moon, rebuilding the world, restoring the FMVs, and preserving the strange Noir soul that made Tex Murphy unforgettable.
The media could not be played.
I just have always used am I crazy? They allow 3 sizes of local model 100% free and have never had an issue
Too lazy to transfer my huge dictionary of slang, names, words, and so on. If anyone can figure out how to do that quickly, I'm all in on trying other dictation projects 
