Hey everyone — I’m the founder of FluidVox, a voice-to-text app for Mac & Windows. I see Wispr Flow recommended constantly in this sub, and for good reason — it’s solid. But I think there’s a gap it doesn’t fill, especially around customization and price. Here’s what I built and why it might be worth your time.
Same concept as Wispr Flow (hold hotkey, speak, get polished text in any app), but with per-app style control, auto-learning corrections, 99 languages, a $39 lifetime local license 14-day FREE Trial No Credit Card required.
-The pricing alone might be enough. Let’s get this out of the way first because it’s the easiest comparison:
–
If you’re happy with Wispr Flow, it’s a great product. But if you want more control over transcription styles per app, auto-learning that adapts to your corrections, or a considerable cheaper option — give FluidVox a try. The free tier is identical so there’s nothing to lose.
Mac and Windows for now, iOS, and Android coming soon.
One more thing — if anyone wants to give FluidVox a proper try beyond the standard 14-day trial, drop a comment and I’ll hook you up with an extended Pro trial. I genuinely want honest feedback, good or bad. Building this has been a great experience and real-world usage feedback is the most valuable thing I can get right now.
Happy to answer any questions here!
Updates:
-
Windows version out now. Please test out the Free trial and let me know if you find any issues. For Windows laptops or desktops highly depends what kind of graphic card you have so test out the local models and see if they work good for you. The cloud version is independent of GPU so that works seamlessly even on low end machines.
Major Updates For MacOS Version: Introducing VOX AGENTWhat Can Vox Agent Actually Do? Real Use Cases
Vox Agent is currently in beta. Everything listed below is functional and actively being improved. Some features are more polished than others, and you may encounter rough edges — that’s expected at this stage.
Vox Agent is a voice-controlled AI assistant that lives in your Mac’s menu bar. You speak naturally, and it chains together 28 tools to get things done — no typing, no clicking through menus. Not sure how to do something? Just ask. Vox can see your screen, understand what you’re looking at, and guide you through step by step.
How It Works
Default Hot Key for the agent:
Option key (hold) — Vox Agent mode. Hold it, speak a command, release — the agent takes over and performs the task. This is how you trigger everything listed below.
Both hotkeys are customizable in settings. The Option key is the default for agent commands, but you can change it to whatever works best for your workflow.
Writing & Text
– Dictate into any app — Hold a key, speak, release. Your words appear wherever your cursor is — email drafts, Slack messages, code comments, Notes, anywhere.- Rewrite selected text — Select a paragraph, tell Vox to “make this more concise” or “rewrite this professionally,” and it replaces the selection in place.- Copy something to clipboard — “Copy my mailing address to clipboard” — ready to paste wherever you need it.
Learning & Guidance
– On-screen guidance — Stuck on anything? Just ask “How do I do this?” or “Where should I click?” Vox takes a screenshot of your screen, analyzes it with vision AI, and walks you through each step — in any app, any workflow. It’s like having someone looking over your shoulder who actually knows what they’re doing.- Learn any app — Trying to figure out Photoshop filters, Excel formulas, or Xcode settings? Just ask while you’re in the app. Vox sees exactly what you see and tells you what to do next.- Create flashcards — “Make flashcards for Spanish vocabulary from chapter 5” → generates an interactive HTML deck with 3D flip animations, keyboard navigation, and progress tracking. Open it in any browser to study.
Email & Messages
– Send an iMessage — “Text Sarah that I’m running 10 minutes late.” Sends via Messages.app.- Send a message with an attachment — “Send John the screenshot I just took” — attaches the latest screenshot automatically.- Check recent conversations — “Show me my recent iMessage conversations” or “What did Mike say in our last chat?”- Search message history — “Find any messages from last week about the dinner reservation.
Calendar & Reminders
– Create events — “Schedule a team meeting for Thursday at 2pm for one hour.”- Check your schedule — “What do I have tomorrow?” or “Am I free Friday afternoon?”- Delete events — “Cancel my 3pm meeting today.”- Create reminders — “Remind me to call the dentist tomorrow at 9am, high priority.”- Check and complete reminders — “What reminders do I have this week?” or “Mark the grocery reminder as done.”
Files & Organization
– Find files — “Search my Documents folder for anything with ‘invoice’ in the name.”- Move and rename — “Move the Q3 report to my Desktop” or “Rename it to Final-Q3-Report.pdf.”- Copy files — “Copy the presentation to my USB drive.”- Clean up — “Delete all the .tmp files in my Downloads folder” — moves them to Trash (recoverable).- Create text files — “Create a markdown file with meeting notes from today” or “Make a Python script that converts CSV to JSON.”
Contacts
– Look up contacts — “What’s Sarah’s phone number?” or “Find everyone at Acme Corp.”- Add new contacts — Just tell the agent the details and it will add to contacts.- Update details — “Update Sarah’s email to -”
Web & Research
– Answer questions — “What’s the capital of Mongolia?” or “How many ounces in a gallon?” — answered directly without opening a browser.- Capture web pages — “Capture the full content of this webpage” — extracts text and images from Safari or Chrome.- Open specific URLs — “Open github in Chrome.”
System & Automation
– Launch apps — “Open Photoshop” or “Switch to Finder.”- Keyboard shortcuts — “Press Command+Shift+4” — triggers any shortcut in the active app.- Run Shortcuts — “Run my ‘Morning Routine’ shortcut” — triggers any macOS Shortcut you’ve built.- Run AppleScript — “Run an AppleScript to set my Mac volume to 50%.” For power users who want custom automation.- Check system status — “How’s my battery?” or “How much storage do I have left?”- Set timers — “Set a timer for 25 minutes” — get a notification when it fires.
Images & Creative
– Generate images — “Create an image of a mountain landscape at sunset” → saves a PNG. Also supports JPG output.- Generate SVGs — “Create an SVG icon of a shopping cart” → static vector graphic, useful for logos, icons, and diagrams. Only generated when you explicitly ask for vector/SVG format.
Documents & Spreadsheets
– Create a PDF from scratch — “Create a PDF invoice for Acme Corp, 3 hours of consulting at $150/hour” → generates a formatted, professional PDF saved to your documents.- Fill out a PDF form — Open a tax form or application in Preview, say “fill in my name, address, and date of birth” and it populates the fields.- Annotate a PDF — “Highlight the third paragraph on page 2” or “Add a note on page 5 saying needs revision.”- Merge PDFs — “Merge these three PDFs into one document.”- Build a spreadsheet — “Create a spreadsheet tracking my monthly expenses for 2026 with totals at the bottom” → generates an .xlsx with formulas and formatting.- Edit a live spreadsheet (Beta) — With Numbers or Excel open: “Add a column for tax at 8.5% and a sum row at the bottom.” It reads your sheet and modifies it in place. Works best with Numbers; Excel support is limited to basic operations for now.- Summarize spreadsheet data — “Summarize this sales data” → generates an HTML report with key insights from your open spreadsheet.- Chart your data (Beta) — “Add a bar chart comparing Q1 vs Q2 revenue” directly in your open Excel spreadsheet. Chart editing is functional but still being refined.
Presentations
– Create a slide deck — “Make a 10-slide pitch deck about our new product launch” → builds slides in Keynote or PowerPoint with structured content.- Edit slides in a live presentation — “Change the title on slide 3 to Q3 Results” or “Add a new slide after slide 5 with our team structure.”- Add speaker notes — “Add speaker notes to slides 1 through 5 with talking points.”- Reorder and clean up — “Move the summary slide to the end” or “Delete slides 7 and 8.”- Export to PDF — “Export this presentation as a PDF for sharing.”
Chaining It All Together
The real power is that Vox chains these tools in a single voice command:
– “Read my open spreadsheet, create a PDF summary with charts, and email it to my boss” → reads the data, generates the PDF, and sends it via Messages.- “Check my calendar for tomorrow, find any conflicts, and text my wife if I have a late meeting.”- “Take a screenshot of this webpage, create flashcards from the content, and save them to my Study folder.”- “Open the Q3 presentation, add a new slide with this month’s sales numbers from the spreadsheet, and export it as a PDF.”- “I’m trying to create a pivot table but I’m stuck” → Vox screenshots your spreadsheet, sees where you are, and walks you through each step.
A Note on API Usage for Free-Key Users
Voice transcription works great even with Google’s free API key — it’s lightweight and stays well within free-tier limits. The Vox Agent, however, is a different story. Agent tasks like generating documents, analyzing screenshots, creating presentations, and multi-step workflows consume significantly more tokens than simple transcription. With heavy use, Google may throttle or limit your free API key. If you run into agent tasks failing or returning errors, your free API key quota is likely the cause. You have two options:
-
Subscribe to Fluid Vox Pro — Everything works out of the box. No API key needed, no usage limits to worry about. All agent features route through our cloud service.
-
Upgrade to a paid Google API key — If you prefer using your own key, a paid tier from Google AI Studio removes the free-tier throttling and lets you use all agent features without interruption.
Transcription will keep working either way — this only affects the heavier agent workflows.
We Need Your Feedback
This is a beta release. Every feature listed above will continue to be improved, and what we work on next depends heavily on what you tell us. If something doesn’t work as expected, if a workflow feels clunky, if you have ideas for new use cases — we want to hear all of it.
– In-app: Go to Settings → Contact Us- On the forum: Post directly in this thread or start a new one
No feedback is too small. Bug reports, feature requests, workflow suggestions, things that confused you — all of it helps us make Vox Agent better. Thank you for being part of this early on.
