Dear OpenAI Team,
First of all, thank you for your incredible work on the voice and language capabilities of ChatGPT. As a native Chinese speaker and a daily user of the voice interface, I truly appreciate the product’s fluency, responsiveness, and emotional nuance.
However, I’ve noticed some minor but important pronunciation issues with Chinese polyphonic characters (多音字) that I’d like to kindly bring to your attention. For example:
-
倔强 (jué jiàng) is sometimes pronounced as jué qiáng, which is incorrect.
-
调度 (diào dù) is occasionally read as tiáo dù, also incorrect in this context.
These are subtle distinctions that greatly affect the naturalness and professionalism of spoken Chinese. I fully understand that Chinese is a highly context-dependent and complex language, especially with so many homophones and tonal variations. That’s exactly why I believe further refinement in polyphonic disambiguation would bring even greater value to users.
I’m not pointing this out as a criticism, but rather as someone who genuinely respects and supports what you’re building. I offer this feedback in the spirit of improvement and with sincere admiration for your work. I would love to see your model grow even stronger and more accurate, especially for a language as rich as Chinese.
Warmest regards,
A loyal user & language enthusiast