sentence with "/" has a bug.

[liuyiheng001]

Self Checks

• This template is only for bug reports. For questions, please visit Discussions.
• I have thoroughly reviewed the project documentation (installation, training, inference) but couldn't find information to solve my problem. English 中文 日本語 Portuguese (Brazil)
• I have searched for existing issues, including closed ones. Search issues
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）
• Please do not modify this template and fill in all required fields.

Cloud or Self Hosted

Self Hosted (Docker)

Environment Details

ubuntu=22.04，python=3.10.15, torch=2.4.1, torchaudio=2.4.1, cuda=11.8

Steps to Reproduce

1. Follow the three steps outlined in docs/zh/inference.md for command-line inference, with --text set to "数字人 (Digital Human / Meta Human) 数字人 (Digital Human / Meta Human) 数字人 (Digital Human / Meta Human)"

2. From the below log image，we can see Encoded text: 数字人（Digital Human / Meta Human）数字人（Digital Human / Meta Human）数字人（Digital Human / Meta Human）
   

3. But the generated audio is missing the last sentence: "数字人 (Digital Human / Meta Human)".

✔️ Expected Behavior

The generated audio should not miss the last sentence: "数字人 (Digital Human / Meta Human)".
I wish you can tell me why this bug is occur?
And hope you can fix this bug.
Thank you very much for your hard work.

❌ Actual Behavior

The generated audio is missing the last sentence: "数字人 (Digital Human / Meta Human)".

[Stardust-minus]

Special characters should be avoided, it will caused model output not stable.