Speech-to-Text

Dictate prompts instead of typing. Built-in Apple speech recognition.

Chapeta supports voice input through Apple's speech recognition APIs. Click the microphone button, speak naturally, and Chapeta transcribes your prompt in real time. It's useful when your hands are busy, when you think faster out loud, or when typing long prompts feels slow.

How to Set Up

  1. Open Chapeta and focus the prompt input.
  2. Click the microphone button beside the input field.
  3. Grant microphone and speech recognition permissions when macOS prompts you.
  4. Start speaking. Your words are transcribed directly into the prompt box.
  5. Edit if needed, then press Enter to send.

What You Get

Live Transcription

As you speak, Chapeta converts speech to text and inserts it into the prompt field in real time.

Voice + Edit

Dictate quickly, then make precise edits with keyboard before sending. Best of both speed and control.

Hands-Free Prompting

Useful for brainstorming, drafting, and quick questions when typing is inconvenient.

Technical Details

Speech input uses Apple's SFSpeechRecognizer and AVAudioEngine. Audio is captured with user permission, transcribed, and inserted locally into the prompt field. Chapeta only sends the final prompt text you submit to the selected model provider.

Requirements

  • macOS 14 (Sonoma) or later
  • Microphone permission (System Settings > Privacy & Security > Microphone)
  • Speech Recognition permission (System Settings > Privacy & Security > Speech Recognition)

FAQ

Try Speech-to-Text with Chapeta