How to Create and Use Personal Voice on iPhone for Natural Voice Synthesis

Losing the ability to speak due to medical conditions—such as ALS (Lou Gehrig’s disease), vocal cord surgery, stroke, or throat cancer—can feel devastating to personal identity. Starting with iOS 17, Apple introduced a groundbreaking accessibility tool called Personal Voice. Powered by on-device machine learning, Personal Voice allows users to record 150 short training phrases to generate a private, remarkably natural digital replication of their voice that can speak typed text aloud during phone calls and FaceTime.

What Is Personal Voice and How Does It Work?

Personal Voice bridges speech preservation with on-device artificial intelligence:

• 100% On-Device Privacy: All audio voice training and neural synthesis occur entirely locally on your iPhone’s Neural Engine while charging overnight. Your personal voice data is never uploaded to Apple cloud servers.
• Live Speech Integration: Integrates directly with iOS Live Speech. Triple-clicking your iPhone side button lets you type phrases to speak them aloud in your custom voice during in-person conversations.
• Phone and FaceTime Ready: Seamlessly speaks your typed responses over active cellular phone calls, FaceTime audio, and supported third-party assistive communication apps.
• Ecosystem Synchronization: Securely synchronizes your trained voice profile across your iPad and Mac using end-to-end encrypted iCloud keychain.

Step 1: Record Your Voice Training Phrases

Setting up Personal Voice takes approximately 15 to 20 minutes in a quiet environment:

1. Ensure your iPhone is updated to iOS 17 or later.
2. Open the Settings app and tap on Accessibility.
3. Scroll down to the Speech section and tap on Personal Voice.
4. Tap Create a Personal Voice.
5. Authenticate using Face ID or Touch ID, then read through the introductory guidance.
6. Find a completely quiet room without background fans, road noise, or echoing walls. Hold your iPhone approximately 6 inches (15 cm) from your mouth.
7. Follow the on-screen audio test to verify your room’s acoustic quality.
8. Begin the recording session: read aloud each of the 150 short phrases displayed on screen (such as casual conversations, nature descriptions, and everyday sentences).
9. Take your time; you can pause the session anytime and resume recording later.

Step 2: Generate the Synthetic Voice Model

Once you complete all 150 phrases, your iPhone must process the neural acoustic model:

1. Plug your iPhone into its charger.
2. Lock the display and connect to Wi-Fi.
3. Voice processing occurs exclusively in the background while your phone is locked and charging overnight.
4. Depending on your iPhone model (iPhone 15 Pro, iPhone 16, etc.), training typically finishes within a single night. You will receive a notification stating “Your Personal Voice is ready”.

How to Speak Using Your Personal Voice

To use your new synthetic voice during daily communication:

1. Go to Settings > Accessibility > Live Speech and toggle it ON.
2. Under Voices, tap on your newly created Personal Voice to select it.
3. Whenever you are on a phone call or in a room with people, triple-click the physical side button.
4. Type your message into the Live Speech box and tap Send. Your iPhone will speak your exact words aloud in your own distinctive voice.