CHINESE SUBJECT PRODUCT

AI Chinese Oral Assessment

Cantonese + Putonghua — live speaking practice

HK-tuned AI oral coach for Chinese: Cantonese and Putonghua natively trained (not a Mandarin model with a Cantonese voice skin). Built-in TSA / DSE oral, recitation, picture-based storytelling, SEN-friendly dialogue. At partner schools every student speaks with AI ~4×/month; weaker students average +59% score gain.

TeachersStudents
CantonesePutonghuaTSA oralDSE oralRecitationStorytellingSENP3–P6S1–S6

Cantonese + Putonghua natively trained · ~4 AI convos/student/month · weaker students +59%

HK TSA Oral Practice
Prompt

“Tell us about a hobby you enjoy.”

0:000:11
Overall
4.2/ 5
Clear HK CantoneseNatural pace
Pronunciation
4.3/5
Fluency
4.0/5
Content
4.5/5
Vocabulary
4.1/5
Add more examples
Transcript

“你好,我最鍾意嘅活動係打籃球,每個星期六都會同同學一齊打。”

Weaker students +59%
Cantonese + Putonghua native

Chinese teacher pain · Essai solution

30-student class — teachers can’t give 1-on-1 Cantonese oral practice to each kid.

AI as personal practice partner. At partner schools, every student speaks with AI ~4 times a month.

Generic AI voice tools only handle Mandarin — Cantonese-native HK students have nowhere to practise.

Cantonese natively trained on HK usage, TSA / DSE prompts, colloquial + book-style vocabulary.

SEN and weaker students avoid speaking in class — end up practising the least.

Private AI dialogue with SEN pacing, repeat prompts, forgiving feedback. Weaker students +59% gain.

Key features

Cantonese + Putonghua native

Each language trained independently. HK vocabulary and pronunciations — not a Mandarin model with a Cantonese voice skin.

TSA · DSE · interview scenarios

Built-in simulations: TSA read-aloud, DSE group discussion, secondary-school and university Chinese interviews.

Recitation practice

Per-phrase scoring for pronunciation, pacing, and tone. AI demo playback for student comparison.

Picture-based storytelling

AI shows a picture and guides students to describe the scene and build a narrative, with expression feedback.

Real-time dialogue · SEN support

Genuine back-and-forth conversation. SEN mode: slower pace, repeat prompts, extended response windows.

School-based prompt bank

HK-curriculum prompts plus teacher-uploaded school-based questions and marking criteria.

How it works

1

Pick the scenario

Teacher assigns a prompt or students pick — TSA / DSE simulation, recitation, picture storytelling, or free dialogue.

2

Student talks with the AI

AI responds live in Cantonese or Putonghua. Pronunciation, fluency, content, and vocabulary scored in real time.

3

Instant feedback + class report

Student sees scored feedback immediately; teacher gets a class weakness heatmap and can export a Word report.

Language + grade coverage

 P3P4P5P6S1S2S3S4S5S6
Cantonese oral
Putonghua oral
Recitation (both)
Picture storytelling
DSE Chinese oral simulation
+59%

Average score gain for weaker students after 8+ weeks of practice · Cantonese + Putonghua native · ~4 AI convos/student/month

HK primary Chinese teacher

“Clean UI across desktop, tablet, and mobile. Content draws heavily from TSA — great pre-exam practice for improving students’ oral skills.”

— HK primary school Chinese teacher

FAQ

Is the Cantonese actually HK-authentic?
Yes. Natively trained on HK Cantonese (mixed colloquial and book-style: 畀 / 嘅 / 你同佢). Not a Mandarin model with a Cantonese voice skin.
What does SEN support include?
Slower speech rate, keyword repeat prompts, extended response windows, forgiving feedback, optional picture cues. Teachers can configure per student.
How is scoring broken down?
Per dimension: pronunciation, fluency, content, vocabulary, dialogue logic. Teachers can customise weights or upload the school’s own marking criteria.
Can we use our school’s own oral prompts?
Yes. Upload or type them in; AI adapts scoring and follow-up questions accordingly.
How was the "+59%" measured?
Students using Essai AI Oral 8+ weeks; pre/post in-school oral scores compared. Weaker-ability group averaged +59% gain; higher groups saw smaller gains (higher starting point).
Is student audio used to train AI models?
No. Audio is used only for that session’s scoring. Each school’s data is isolated; contract forbids training on school data.
Can we fund this with the EDB HK$500k programme?
Yes. Essai meets EDB Circular 221/2025. Book a strategy session to get the application template.

AI Tutor

General chat

Reply

How can I help?

Ask me anything or pick a suggestion below.