How to Configure and Use Live Speech and Personal Voice on iPhone for Accessibility

For individuals facing progressive medical conditions that impact speech—such as ALS (Amyotrophic Lateral Sclerosis), stroke recovery, or vocal cord trauma—preserving vocal identity is deeply meaningful. In iOS 17 and later versions (including iOS 18), Apple introduced two transformative neural accessibility features: Personal Voice and Live Speech. Powered by the on-device Apple Neural Engine, Personal Voice trains a synthesized replica of your unique vocal timber from a series of audio prompts, while Live Speech allows you to type phrases that are spoken aloud in your custom voice during phone calls, FaceTime sessions, and in-person conversations.

Understanding Personal Voice vs. Live Speech

The two features work together as an integrated vocal accessibility pipeline:

• Personal Voice: The local machine learning model trained on your voice. Processing takes place exclusively on your iPhone while connected to power and locked, guaranteeing complete biometric cryptographic privacy.
• Live Speech: The typing interface. Whenever you triple-click the Side button or Back Tap, a text input card appears. Anything you type is spoken aloud through the iPhone speakers or piped directly into cellular and FaceTime audio streams.

Step-by-Step: Training Your Personal Voice on iPhone

Training a Personal Voice requires approximately 15 to 20 minutes in a quiet indoor room:

1. Access the Personal Voice Settings

1. Open the Settings app on your iPhone.
2. Scroll down and tap Accessibility.
3. Under the Speech section, tap Personal Voice.
4. Tap Create a Personal Voice and authenticate using Face ID or your passcode.

2. Record the Audio Prompts

1. Hold your iPhone approximately 6 inches (15 cm) from your face.
2. Follow the on-screen instructions to read 150 distinct phrases aloud (taking roughly 15 minutes).
3. If ambient background noise interferes with a recording, iOS will prompt you to re-read the sentence.

3. On-Device Machine Learning Synthesis

Once recording is finished, your iPhone must generate the acoustic neural model. Because model generation is computationally intensive, iOS processes Personal Voice only when your iPhone is connected to a charger and locked. Typically, the voice model finishes generating overnight.

Enabling and Using Live Speech

Once your Personal Voice is created—or if you prefer using Apple’s high-quality system voices (like Siri)—activate Live Speech:

1. Go to Settings > Accessibility > Live Speech.
2. Toggle the main Live Speech switch to On.
3. Under Voices, select Personal Voice and choose your synthesized profile.

Speaking Aloud in Conversations:

• Triple-Click the Side Button: Quickly press the physical Side button on your iPhone three times to reveal the Live Speech keyboard card.
• Type to Speak: Type your sentence and tap the blue Send/Speak button. The iPhone will articulate your words in real time.
• During Phone Calls and FaceTime: The synthesized speech streams directly to the call recipient with zero acoustic feedback.

Saving and Organizing Favorite Quick Phrases

For everyday interactions (such as ordering coffee, greeting family, or stating emergency details), typing out full sentences is slow. Live Speech lets you build a bank of pre-saved phrases:

1. In Settings > Accessibility > Live Speech, tap Favorite Phrases.
2. Tap the + button in the top-right corner.
3. Type commonly used expressions (e.g., “Hello, how are you today?” or “I need a glass of water, please.”).
4. When Live Speech is active, tap the Phrases tab to speak any saved sentence with a single tap.