26 seconds of on-device AI on iPhone — does the result screen make sense?

I made a short walkthrough of 他爱我吗 (Does He Love Me?), my Chinese-language iOS app:

https://x.com/yang9527pf/status/2104726395943600248

The idea: ask a relationship question, add some context, and compare possible answers instead of getting one confident-sounding response.

Laya runs locally through Core ML. Chinese speech transcription and history stay on the phone too. No account or cloud inference.

The part I’m still thinking about is the result screen. Percentages are easy to read—but also easy to overread. These are relative model scores, not measured probabilities of someone’s feelings.

After watching the demo, what would you change first?

• The wording of the answers

• How the scores are displayed

• How the “uncertain” option is explained

A specific confusing moment would be especially useful feedback. No need to buy the app to comment on the walkthrough.

I’m the maker. The current app has a Chinese interface and is a one-time paid download.

English overview: https://doesheloveme.tennote.site/en/

Disclosure: this update was drafted with AI assistance.