# Configuring Transcription

Set up offline Apple Intelligence or Cloud APIs for speech-to-text.

Before recording, you need to select the **service provider** that will process and transcribe your audio. You can access these settings via the **Configure** button in the Recording sidebar tab.

### [Option 1: Apple Intelligence (Offline & Free)](/content/docs/configuring-transcription#option-1-apple-intelligence-offline--free/index.html)

This uses the on-device speech recognition capabilities built directly into macOS. It is completely private, free, and works without an internet connection.

1. In the Recording view, click **Configure**.
2. Select **Apple Intelligence**.
3. **Manage Languages**: You will see a status list of supported languages.
   - **If the language is "Installed":** You are ready to go.
   - **If the language is "Not Installed":** Click the download icon next to the language to install the necessary pack to your device.

#### [Supported Languages](/content/docs/configuring-transcription#supported-languages/index.html)

Silkwave leverages Apple Intelligence ( **macOS 26+**) to transcribe the following languages on-device:

- Cantonese 🇭🇰
- Chinese 🇨🇳
- English 🇺🇸
- French 🇫🇷
- German 🇩🇪
- Italian 🇮🇹
- Japanese 🇯🇵
- Korean 🇰🇷
- Portuguese 🇵🇹
- Spanish 🇪🇸

### [Option 2: OpenAI API (Cloud)](/content/docs/configuring-transcription#option-2-openai-api-cloud/index.html)

For additional languages not supported by Apple Intelligence, or to use specific cloud models, you can connect to the OpenAI API.

1. In the Recording view, click **Configure**.
2. Select **OpenAI**.
3. Enter your **API Key** (if you haven't already set it up in the Chat section).
4. **Select Model**: Choose from the available OpenAI models (e.g., gpt-4o-transcribe) for transcription.