Settings & Preferences

Every toggle and option in AsPeach, explained.

Open Settings by clicking the gear icon in the top corner of the AsPeach window.

General Settings

These are the main toggles that control AsPeach's behavior:

Microphone

Selecting Your Microphone

The Microphone dropdown lists all available audio input devices on your computer. Select the one you want to use for recording.

• Default — Uses whatever your system's default microphone is.
• If the selected mic becomes unavailable (e.g., you unplug a USB mic), AsPeach automatically switches to another available mic.

Testing Your Microphone

Click the Test Mic button. A live waveform appears showing your audio input in real-time for 5 seconds. Speak or clap to see the waveform react. This confirms:

• The correct mic is selected.
• Your mic is actually picking up sound.
• The volume level is reasonable.

Whisper Models

Whisper is the AI engine that converts your speech to text. It comes in different sizes — bigger models are more accurate but take longer to process.

Which Model Should I Use?

• Apple Silicon Mac (M1/M2/M3/M4): Use Medium or Large. Your hardware handles them easily.
• Intel Mac: Use Base or Small. Larger models may be very slow.
• Windows: Depends on your CPU. Start with Small and try Medium if performance is acceptable.

Downloading Models

When you select a new model, AsPeach downloads it automatically. The download happens once — after that, the model is cached locally on your computer.

Languages

Setting Your Languages

1. Go to the Languages section in Settings.
2. Use the search box to find and add languages you speak.
3. Your primary language is the first language in the list. This is used when auto-detect is turned off.
4. Drag languages to reorder them (move your primary language to the top).

Auto-Detect vs. Fixed Language

• Auto-Detect ON (default) — Whisper figures out which language you're speaking in each recording. Great if you switch between languages.
• Auto-Detect OFF — Always transcribes using your primary language. This can be more accurate if you only ever speak one language, because Whisper doesn't have to guess.

AsPeach supports 40+ languages out of the box. You can switch languages mid-sentence and Whisper will usually handle it.

LLM Setup (for AI Features)

AI features like transforms, auto-edit tone adjustment, and context-aware pasting require a Large Language Model (LLM). AsPeach supports three providers:

Option 1: Ollama (Default — Free, Local, Private)

Ollama runs AI models entirely on your computer. No internet connection needed, no API costs, and complete privacy — nothing leaves your machine.

1. Download and install Ollama from ollama.ai.
2. Open your Terminal (Mac) or Command Prompt (Windows) and type: ollama pull llama3.2
3. Wait for the model to download (this may take a few minutes depending on your internet speed).
4. In AsPeach, go to Settings → LLM Provider and select Ollama.
5. Select the model you just downloaded from the dropdown.
6. A green dot appears next to the dropdown if Ollama is connected and ready.

Recommended models: llama3.2 (3B parameters), llama3.1 (8B), mistral (7B). Models smaller than 3B parameters may produce poor results — AsPeach will show a warning if you select one.

Option 2: OpenAI (Cloud, Requires API Key)

1. Create an account at platform.openai.com and generate an API key.
2. In AsPeach Settings → LLM Provider, select OpenAI.
3. Paste your API key in the field that appears.
4. Select a model from the dropdown (e.g., gpt-4o-mini).

Note: OpenAI charges per usage. Your audio is still processed locally by Whisper — only the text transform step uses the cloud API.

Option 3: Anthropic (Cloud, Requires API Key)

1. Create an account at console.anthropic.com and generate an API key.
2. In AsPeach Settings → LLM Provider, select Anthropic.
3. Paste your API key in the field that appears.
4. Select a model from the dropdown (e.g., claude-sonnet-4-6).

Status Indicator

A colored dot next to the LLM model dropdown shows connection status:

• Green — Connected and ready.
• Red — Not connected. Check that Ollama is running, or that your API key is correct.

You can switch providers anytime without restarting AsPeach.

Keyboard Shortcut

The keyboard shortcut is how you start and stop recording from anywhere on your computer — even when AsPeach is in the background or minimized.

Default Shortcut

• Mac: Option + Shift + Space (⌥⇧Space)
• Windows: Alt + Shift + Space

How to Change It

1. In Settings, find the Keyboard Shortcut section.
2. Click the Record Shortcut button. It changes to say "Press keys..."
3. Press the key combination you want. For example, hold Ctrl + Shift and press R.
4. The new shortcut is saved immediately and displayed on screen.

Conflict Detection

If the shortcut you choose is already registered by another application, AsPeach will show a warning. Pick a different combination to avoid conflicts.

Per-App Shortcuts

You can assign different keyboard shortcuts for different applications. For example, use one shortcut in your browser and a different one in your email client. Set this up in the Per-App Context section of Settings.

Other Keyboard Shortcuts

• ⌘Z / Ctrl+Z — Undo the last AI transform (within a 30-second window after it was applied).
• Escape — Close the preview panel or close settings.