How to Convert Text to Speech on iPhone

Q: How much text can I convert at once?

Up to 4,000 characters per generation — roughly 600-700 words, or 4-5 minutes of speech at normal rate. For longer material, split it into sections and generate each; the history keeps them in order.

Q: Does text-to-speech work offline?

Generation needs a connection, since the AI runs in the cloud. Everything you've already generated is stored on your device and plays back offline — generate before the flight, listen in the air.

Q: Can I change the voice or speed after generating?

Playback speed, yes — 0.5x to 2x anytime. The voice and speaking rate are baked in at generation, so to change them, regenerate the text with new settings; your history keeps both versions.

iPhones have read text aloud for years — in a voice nobody would choose on purpose. Modern AI voices are a different species: intonation, pauses, personality. Here's the full workflow from typed text to a natural-sounding audio file.

The one-minute version

Open Voiceify and type (or dictate) your text into the composer — up to 4,000 characters, with a live counter.
Tap Voice and audition the options — all 13 have tap-to-play previews and personality tags. Cedar and Marin are the recommended defaults.
Set the Speaking Rate — six steps from Very Slow to Very Fast — and a Response Format (MP3 unless you have a reason; see the formats guide).
Tap Convert to Speech. Seconds later the audio is playing, with a scrubber, ±15-second skips and 0.5–2× playback speed.
Share it anywhere or Save to Files — and it's already in your history regardless.

Getting better-sounding results

Punctuation is direction. AI voices read commas as breaths and periods as full stops. Write the pauses you want to hear; an em-dash buys you a dramatic beat.
Spell it like it sounds. Acronyms, unusual names and numbers sometimes read better phonetically — “S-Q-L” vs “sequel” is your call to make in the text.
Match voice to material. A bedtime story in Onyx sounds like a documentary; the voice guide pairs all 13 with their best material.
Rate at generation beats speed at playback. If you know you want it slow (language practice, instructions), generate at Slow rather than slowing playback — pacing sounds more natural.

Working with longer text

4,000 characters is about 600–700 words — four to five minutes of speech. For longer material, split at natural section breaks and generate each part. The history keeps every generation with its title, duration and format, searchable and renamable, so a chaptered document stays organized. Generations live on your device and play back offline.

Honest note: all Voiceify output is AI-generated speech — lifelike, but synthetic, and the app is upfront about that. If you're publishing the audio, disclose it where platforms require.

Hear your first text in seconds

13 previewable voices, 6 rates, 4 export formats. Free to try on iPhone & iPad.

Download on the App Store

FAQ

How much text can I convert at once?

4,000 characters — roughly 600–700 words. Split longer material into sections; history keeps them ordered.

Does text-to-speech work offline?

Generation needs a connection; everything in your history plays back offline from your device.

Can I change the voice or speed after generating?

Playback speed anytime (0.5–2×); voice and speaking rate require regenerating — history keeps both versions.