AI runs 100% on your iPhone — your voice never leaves the device
Understand English pronunciation — don't just read for an AI to score
Vocula shows you what your tongue, lips, and vocal folds must do for every sound, measures your voice with real acoustics, then diagnoses in plain language: what went wrong and how to fix it — not a vague percentage.
The problem with scores
"78% match" doesn't teach you how to fix it
You read a sentence, the machine returns a number. You know you're wrong — but not where, and certainly not what to do with your tongue. Vocula works the other way: observe, diagnose, fix.
Typical scoring app
"78% match. Try again."
A number from a black box. Your tenth attempt is still 78%, because you don't know what to change.
Vocula
"/θ/ — tongue tip didn't reach the front teeth. Place the tip between your teeth, push air out, keep vocal folds still."
A diagnosis from real acoustic measurement, with concrete mechanical guidance so the next attempt differs from the last.
Features
See the sound, fix the mechanics
/θ/
Animated articulation diagrams
Each of the 44 GA + RP sounds has a sagittal cross-section of the mouth: tongue position, jaw opening, lip shape, airflow, and voicing. You see exactly what your articulators must do.
F1·F2
Articulation gym
Real-time biofeedback drills: lower your F3 to master /r/, separate /s/–/ʃ/–/θ/ by spectral shape, and train perception with evidence-backed HVPT listening drills.
0 server
Absolute privacy
Whisper and Wav2Vec2 run as Core ML models directly on your iPhone. No scoring server, no audio uploads, fully offline practice.
Learning path
Three tiers — from single sounds to real speech
Tier 1
Phonemes
Master individual sounds with articulation diagrams, formant and spectral measurement. The foundation everything else builds on.
Tier 2
Prosody
Sentence stress, rhythm, and pitch contour — the musicality of English, the widest gap for most learners.
Tier 3
Shadowing
Apply everything to real passages with the shadowing method, scored by explainable per-phoneme assessment.
Science & honesty
Real measurement, honest feedback
- Research-basedDrills are designed from peer-reviewed phonetics and pronunciation-training research — HVPT, formant biofeedback, interleaved practice against rote memorization.
- No fake numbersWhen there isn't enough data, Vocula says "not enough data" instead of showing a number that merely looks nice. Mastery is tracked per phoneme.
- Transparent pricingNo auto-renew traps, no murky charges. The core phoneme trainer is free forever, no account required.
Pricing
Free to start, transparent to upgrade
Free
$0
Core phoneme training with articulation diagrams and acoustic feedback. No account, no ads.
Pro
Announced at launch
All 44 GA + RP sounds, the articulation gym, prosody, shadowing, and the mastery dashboard. Clear list price — no renewal traps.
FAQ
Frequently asked questions
How is Vocula different from Elsa Speak and other scoring apps?
Scoring apps return a percentage and ask you to try again. Vocula shows you the mechanics that produce each sound — tongue, lips, vocal folds — measures your voice with real acoustic quantities (formants, spectrum, pitch contour) and gives a plain-language diagnosis: what went wrong and how to fix it.
Is my voice sent to a server?
No. All AI (Whisper and Wav2Vec2 as Core ML models) and every acoustic measurement run directly on your iPhone. Audio never leaves the device and there is no scoring server.
Do I need an internet connection to practice?
No. Once lessons and models are downloaded, practice works fully offline.
Does Vocula teach American or British English?
Both. Vocula's IPA system supports General American (GA) and Received Pronunciation (RP) side by side, with symbols normalized consistently across all lessons.
When does Vocula launch?
Vocula is in final polish and will launch on the App Store soon. Check back here for updates.
Who is Vocula for?
Serious learners who want to understand and fix their pronunciation at the root: IELTS Speaking 7.0+ candidates, professionals who work in English, and teachers who want a visual phonetics teaching tool.