Voice Input
Speak your ideas naturally and let Prompt Assistant transcribe and improve them automatically. No typing required.
Pro Feature: Voice input is available exclusively for Pro subscribers.Upgrade to Pro
Requirements
- Pro subscription - Voice input is a Pro-only feature
- Microphone permission - Grant access in System Settings or during onboarding
- Working microphone - Built-in or external
How It Works
- Record - Speak your prompt idea naturally
- Transcribe - Whisper AI converts your speech to text
- Improve - The transcribed text is automatically improved
- Use - Copy the polished prompt
Using Voice Input
Method 1: Keyboard Shortcut
- Press Cmd+Shift+U from anywhere
- The popover opens and recording starts automatically
- Speak your prompt idea
- Press Cmd+Shift+U again or click Stop
- Wait for transcription and improvement
- Copy your improved prompt
Method 2: Microphone Button
- Open the Prompt Assistant popover
- Click the microphone icon in the input area
- Speak your prompt
- Click the stop button when done
- The text appears and is improved automatically
Recording Interface
While recording, you'll see:
- Waveform visualization - Shows audio levels in real-time
- Timer - Displays recording duration (MM:SS)
- Stop button - Red button to end recording
- Cancel button - X button to discard the recording
Tips for Better Results
Speaking Clearly
- Speak at a normal pace - not too fast or slow
- Enunciate clearly, especially technical terms
- Minimize background noise
- Position microphone at appropriate distance
Structuring Your Thoughts
- Think about what you want to say before recording
- Start with the main idea, then add details
- It's okay to pause briefly - the transcription handles it
- You can re-record if needed
What to Say
You don't need perfect grammar or structure. Just express your idea:
Example
You say: "I need a video of like, a sunset over the ocean, with maybe some birds flying, and it should feel peaceful and calm"
You get: A polished video prompt with camera movements, lighting details, and technical specifications
Transcription Technology
Voice input uses OpenAI Whisper for transcription:
- Supports multiple languages
- High accuracy even with accents
- Handles technical terminology well
- Processes in a few seconds
Troubleshooting
Microphone Not Working
- Check System Settings > Privacy & Security > Microphone
- Ensure Prompt Assistant has permission
- Try a different microphone if available
- Restart the app after granting permission
Poor Transcription Quality
- Reduce background noise
- Speak closer to the microphone
- Speak more slowly and clearly
- Check your internet connection
Feature Not Available
- Verify you have an active Pro subscription
- Check Settings > Account for subscription status
- Try signing out and back in
Best Use Cases
Brainstorming
Quickly capture ideas without stopping to type. Great for:
- Creative concepts for images/videos
- Rough drafts of messages
- Quick notes that need polishing
On the Go
Record prompts when typing isn't convenient:
- Walking or commuting
- When hands are busy
- Quick captures between tasks
Long-Form Content
Speaking is faster than typing for longer prompts:
- Detailed video scene descriptions
- Complex image compositions
- Multi-paragraph text requests
Customizing the Shortcut
You can change the voice input keyboard shortcut:
- Open Settings > Shortcuts
- Find "Voice Recording"
- Select a new key combination
Default is Cmd+Shift+U