The audio generated for Japanese and Korean is incomplete.

Description

I noticed that when generating audio for Japanese and Korean, the last word of the speech is always incomplete. May I ask what the issue might be?

incomplete.wav

Reproduction

"input": "아, 고맙습니다. 음, 회의 목적에 대해 설명드리겠습니다."

Logs

Environment Information

python3.11

Known Issue

• The issue hasn't been already addressed in Documentation, Issues, and Discussions.