How to Make a Robot Voice Loop on iPhone

A robot vocal loop is a few short words, spoken in a buzzy synthetic voice, repeating in time. Think vintage computer voices over a beat. On iPhone you can build one in a minute with SpellSynth's SEQ mode: sixteen steps, each holding a word.

Open Settings and choose a VOICE CHARACTER. Each one comes from a different era of speech synthesis:

• 1968 Odyssey: a deep, slow, monotone baritone, a nod to late-60s sci-fi computers.
• 1978 LPC: the buzzy speaking-toy sound most people picture.
• 1986 DECtalk Style: deliberate and emphatically computery.
• Default: clean system speech when you just want it clear.

The voice applies to every word, so you can switch later and the whole loop changes character.

1. Open SEQ.
2. Tap a step, type a word, tap DONE. Two or more letters is a word.
3. A single letter is spoken as a letter, so you can spell a word across a few steps and then say it whole on the next one.
4. Leave a step empty for a rest.

Four words is enough to start, for example type it, play it.

Tap ▶ PLAY. BPM runs from 40 to 240, and steps land on quarter, eighth or sixteenth notes. Turn on FIT and each word stretches to fill its step exactly, speeding up and pitching up as you raise the tempo, the way a sampler does. Turn it off and each word plays at its natural speed until the next step cuts it off.

SEQ plays from the 500+ word library when the word is in it, and falls back to on-device speech through your chosen voice when it isn't. Nothing leaves your device. SpellSynth speaks French too: switch VOICE LANGUAGE to Français.

SpellSynth is also an AUv3 instrument. The plugin has the same sequencer and follows your host's tempo and transport, so the loop stays locked to your beat. Two factory presets show the idea: Robot Chant puts a word on every step, Spelling Bee spells a word out and then says it.

More in the SEQ chapter of the manual.