← All Stories
D
David W.
New York, NY · 🇨🇳 Mandarin
Heritage

A1 to A2 Mandarin — a call with his grandmother

V8 Program · 8 months · A1A2

GOAL
Heritage reconnection — grandparents speak no English
"HEXI never let me skip tones. The pronunciation ring kept me honest. At month four I called my grandmother for the first time in Mandarin. She cried. So did I."
160
Sessions
96h
Voicely Hours
192h equiv
Classroom equiv
820
Words protected

VOICECAST™ Snapshot

Day 0A1
22%PRON
28%COMP
19%PROD
25%RET
+45pts avg
Day 240A2
68%PRON
72%COMP
61%PROD
74%RET

4-Ring Progress

Pronunciation22% → 68%+46pts
Comprehension28% → 72%+44pts
Production19% → 61%+42pts
Retention25% → 74%+49pts
Overall CEFRA1A2

The Full Story

David's grandparents came to New York from Shandong province in 1971 and never learned English. He had grown up understanding fragments — tones, food words, family names — but had never spoken back.

VOICECAST confirmed the pattern: pronunciation ring at 22%, comprehension at 28%. He understood more than he could say, but both were weak in absolute terms. HEXI assigned a V8 — Mandarin's FSI Category IV multiplier makes 8 months the honest minimum for meaningful A2.

The program was unrelenting about tones. Shadow Mode replayed native audio of each phrase until his pronunciation ring registered the correction. He failed the tone check on 声调 (shēngdiào) eleven times in week two. On week six, he passed it on the first attempt.

Month four: his first full Mandarin phone call with his grandmother. Month eight: he recorded a conversation and played it for his mother. She said the tones were right — not 'good for an American,' but right.

VOICECAST™ History

DayCEFRPRONUCOMPRPRODURETEN
Day 0A122%28%19%25%
Day 60A139%43%33%41%
Day 120A252%56%46%56%
Day 180A261%65%54%67%
Day 240A268%72%61%74%

Phoneme Accuracy Arc

Day 0 — Weak Phonemes
Tone 2Tone 3/ɕ//ʐ//ts/
Day 240 — Remaining
Tone 3
RESEARCH ANCHOR

DeKeyser (2007): Tonal accuracy requires corrective phoneme feedback — not repetition alone. HEXI's Shadow Mode provides the error signal that classroom repetition lacks.

These profiles are representative composites grounded in real CEFR research trajectories and Voicely's 2× efficiency model. Scores are generated using Voicely's actual 4-ring assessment algorithm. Individual results vary based on study consistency, native language, and prior exposure.

Get your own VOICECAST™ free →
14 days free · No credit card
← Back to all stories