OpenAI TTS Configuration

OpenAI provides neural text-to-speech voices. Anx Reader supports OpenAI-compatible TTS endpoints, including official OpenAI and compatible proxies.

Step 1: Create an OpenAI API Key

1. Visit OpenAI API Keys.
2. Sign in and click Create new secret key.
3. Copy the key (it is shown only once).

Step 2: Choose the Endpoint URL

The default OpenAI TTS endpoint is: https://api.openai.com/v1/audio/speech

If you are using a proxy or compatible service, replace the URL with the one provided by your vendor.

Step 3: Choose a Model

Recommended model: gpt-4o-mini-tts.

You can check the latest model list in the OpenAI TTS guide: https://platform.openai.com/docs/guides/text-to-speech

Step 4: Choose a Voice

You can type a voice name manually or select one from the voice list and tap Use to fill the field.

Common voices include:

• alloy
• ash
• coral
• echo
• fable
• nova
• onyx
• sage
• shimmer

Step 5: Configure in Anx Reader

1. Open Settings → Narrate.
2. Set TTS Service to OpenAI.
3. Fill in:
   • URL
   • API Key
   • Model
   • Voice
   • Instructions (optional)

Notes:

• The app appends two extra sentences to your Instructions to apply rate and pitch.
• Use the Test button to validate the configuration.