Documentation

Last updated: March 2, 2026

Everything you need to get started with AsPeach and make the most of voice-to-text on your Mac.

1. Getting Started

1.1 System Requirements

  • macOS 12 (Monterey) or later
  • Apple Silicon (M1/M2/M3/M4) or Intel processor
  • 4 GB RAM minimum (8 GB recommended for larger Whisper models)
  • 500 MB disk space for the application and default model

1.2 Installation

  • Download AsPeach from the download section on our homepage.
  • Open the .dmg file and drag AsPeach into your Applications folder.
  • Launch AsPeach from Applications or Spotlight.
  • On first launch, macOS may ask you to confirm opening the app — click Open.
  • Grant Accessibility permission when prompted (required for text insertion).
  • Grant Microphone permission when prompted.

1.3 First Transcription

  • AsPeach automatically downloads the small Whisper model on first launch.
  • Press the global hotkey (default: Option + Space) to start recording.
  • Speak naturally into your microphone.
  • Press the hotkey again (or release, depending on your mode) to stop.
  • Your transcribed text appears wherever your cursor is placed.

2. Core Features

2.1 Voice-to-Text

AsPeach uses OpenAI's Whisper AI model running 100% locally on your Mac. No audio data is ever sent to any server. The transcription happens entirely on your hardware, ensuring complete privacy.

2.2 Whisper Models

Choose the model that fits your needs:

  • Tiny — Fastest, lowest accuracy. Good for quick notes.
  • Base — Balanced speed and accuracy for everyday use.
  • Small (default) — Great accuracy for most users.
  • Medium (Pro) — High accuracy, handles accents well.
  • Large (Pro) — Best accuracy, requires more processing power.

Models are downloaded automatically when selected. Larger models require more RAM and processing time but deliver significantly better accuracy.

2.3 Context-Aware Output (Pro)

AsPeach detects which application you're using and adjusts the tone and formatting of the transcribed text:

  • Slack / Messages — Casual, conversational tone.
  • Email (Mail, Gmail) — Professional, polished tone.
  • VS Code / Terminal — Raw text, no formatting changes.
  • Notes / Word — Clean, well-punctuated output.

Context-aware mode uses a local AI model (Ollama) to intelligently format your speech based on the destination.

2.4 AI Auto-Edit (Pro)

When enabled, AI Auto-Edit automatically cleans up your transcription:

  • Removes filler words ("um", "uh", "like", "you know")
  • Fixes grammar and punctuation
  • Adds proper capitalization and sentence structure
  • Preserves your intended meaning while polishing the output

2.5 Command Mode (Pro)

Use voice commands to control text editing without touching the keyboard:

  • "Delete last paragraph" — Removes the previous paragraph.
  • "Make it formal" — Rewrites the last transcription in a formal tone.
  • "Undo" — Reverts the last action.
  • "New line" / "New paragraph" — Inserts line breaks.
  • "Select all" — Selects all text in the current field.

2.6 Whisper Mode (Pro)

Whisper Mode allows you to dictate at a low volume. The sensitivity is increased so you can speak quietly — perfect for open offices, libraries, or late-night work sessions.

2.7 Custom Dictionary (Pro)

Teach AsPeach your vocabulary:

  • Add brand names, technical terms, and proper nouns.
  • Set phonetic variations (e.g., "aspeach" → "AsPeach").
  • Import/export your dictionary for backup.

2.8 Text Snippets (Pro)

Create voice-activated text snippets for frequently used phrases:

  • Say "my email" → inserts your full email address.
  • Say "signature" → inserts your email signature.
  • Supports multi-line snippets and dynamic variables (date, time).

2.9 Multilingual Support

AsPeach supports 30+ languages out of the box. You can switch between languages mid-sentence — no need to toggle a language setting. Whisper automatically detects the language you're speaking.

3. Keyboard Shortcuts

  • Option + Space — Start/stop recording (default, customizable)
  • Escape — Cancel current recording
  • Cmd + Shift + V — Paste last transcription

You can customize the global hotkey in Settings → Hotkey.

4. Pro Features & Sync

4.1 Cross-Device Sync

Pro subscribers can sync their settings across multiple Macs:

  • Custom dictionary entries
  • Text snippets
  • Application settings and preferences
  • Transcription history

Sync data is stored securely on our servers. Your audio recordings are never synced — only settings and text-based data.

4.2 Device Management

AsPeach Pro allows one active device at a time. When you sign in on a new Mac, the session on your previous device is automatically ended. This ensures your subscription is personal and secure.

5. Privacy & Security

AsPeach is built with privacy as a core principle:

  • All transcription happens locally — your audio never leaves your Mac.
  • No telemetry or tracking — we don't collect usage data from the desktop app.
  • No account required for the free tier — just download and use.
  • Pro sync data is encrypted and stored securely.
  • Read our full Privacy Policy for details.

6. Troubleshooting

6.1 Microphone Not Detected

  • Go to System Settings → Privacy & Security → Microphone.
  • Ensure AsPeach is listed and enabled.
  • Try restarting AsPeach after granting permission.

6.2 Text Not Inserting

  • Go to System Settings → Privacy & Security → Accessibility.
  • Ensure AsPeach is listed and enabled.
  • Some applications may require additional permissions or may not support text injection.

6.3 Poor Transcription Quality

  • Try a larger Whisper model (Medium or Large) for better accuracy.
  • Check your microphone — use an external mic for best results.
  • Reduce background noise in your environment.
  • Speak clearly and at a normal pace.

6.4 High CPU Usage

  • Larger models (Medium, Large) require more processing power.
  • On Intel Macs, consider using the Small or Base model for better performance.
  • Close other resource-intensive applications during transcription.

7. Contact & Support

Need help? Reach out to us: