The generated sound turned into noise or gurgling sounds

[hadestyz]

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

Linux-x86；
docker run -it
--name fish-speech
--gpus all
-p 7860:7860
-v /home/runone/tyz:/home/runone/tyz
fishaudio/fish-speech:latest-dev
zsh

python -m tools.api_server --listen 0.0.0.0:7860 --llama-checkpoint-path "checkpoints/fish-speech-1.5" --decoder-checkpoint-path "checkpoints/fish-speech-1.5/firefly-gan-vq-fsq-8x1024-21hz-generator.pth" --decoder-config-name firefly_gan_vq --compile --half

Steps to Reproduce

It uses a fixed tone, with a WAV file size of 3MB and a duration of about one minute, and is a simple human voice.
I generate Chinese speech, and occasionally when reading English or symbols, it becomes noisy or irregular sounds, even in pure Chinese. My machine performance is sufficient. I am using Nivdia-A30 and the environment should be fine. I directly pulled the Docker image. Is there any way to avoid or reduce similar situations?

✔️ Expected Behavior

Generate smooth speech

❌ Actual Behavior

Occasionally, there may be complete noise or irregular sounds

[hadestyz]

When using the following text, noise will be generated:
"SSH 配置文件：检查 /etc/ssh/sshd_config 是否配置正确。"

It seems to be because the '/' symbol was used