elevenlabs / elevenlabs-docs

Documentation for elevenlabs.io/docs
https://elevenlabs.io/docs
54 stars 240 forks source link

Feedback on Vietnamese Text-to-Speech Feature in Eleven Labs #350

Open Carethi2020 opened 1 month ago

Carethi2020 commented 1 month ago

Dear Eleven Labs Team,

I hope this message finds you well. I am writing to express my appreciation for the Eleven Labs product. As a Vietnamese speaker, I am interested in using it for Vietnamese language processing. However, I have encountered several issues with the text-to-speech feature.

Vietnamese has six tonal marks that significantly affect pronunciation and meaning:

Thanh ngang: No mark (e.g., "ca," "ba") Thanh huyền: Mark ` (e.g., "cà," "bà") Thanh hỏi: Mark ? (e.g., "cả," "bả") Thanh ngã: Mark ~ (e.g., "cã," "bã") Thanh sắc: Mark ´ (e.g., "cá," "bá") Thanh nặng: Mark . (e.g., "cạ," "bạ")

Currently, the system does not accurately recognize these marks, leading to incorrect pronunciations. This mispronunciation can completely alter the meaning of words. For instance, "cá" means fish, "ca" means sing, and "cả" means all. The inaccuracies make the text-to-speech service unusable for Vietnamese.

I hope you can address these issues to improve the functionality for Vietnamese users. If there is any way I can assist in this process as a native speaker with a deep understanding of the language, I am more than willing to help.

Thank you for your attention to this matter. I look forward to seeing improvements that will benefit the 100 million Vietnamese in Vietnam. Best reards, CareThi