xinference plugin: I can't select voice when using tts #111

Closed
opened 2026-02-16 10:18:06 -05:00 by yindo · 13 comments
Owner

Originally created by @937089773 on GitHub (Mar 21, 2025).

Self Checks

  • This is only for bug report, if you would like to ask a question, please head to Discussions.
  • I have searched for existing issues Dify issues & Dify Official Plugins, including closed ones.
  • 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 the required fields.

Dify version

all

Plugin version

0.0.2

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

Image
As shown in the above figure, after I selected the Cosyvoice model, I was unable to select the voice information. This is because the get_customizable_model_schema method in the dify-official-plugins\models\xinference\models\tts\tts.py file did not return the model_properties.
This is the code I fixed.

Image
It can now be displayed normally.

✔️ Error log

No response

Originally created by @937089773 on GitHub (Mar 21, 2025). ### Self Checks - [x] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general). - [x] I have searched for existing issues [Dify issues](https://github.com/langgenius/dify/issues) & [Dify Official Plugins](https://github.com/langgenius/dify-official-plugins/issues), including closed ones. - [x] I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)). - [x] [FOR CHINESE USERS] 请务必使用英文提交 Issue,否则会被关闭。谢谢!:) - [x] Please do not modify this template :) and fill in all the required fields. ### Dify version all ### Plugin version 0.0.2 ### Cloud or Self Hosted Self Hosted (Docker) ### Steps to reproduce ![Image](https://github.com/user-attachments/assets/13a20e14-1db9-4606-8fb9-3e377fd928bf) As shown in the above figure, after I selected the Cosyvoice model, I was unable to select the voice information. This is because the get_customizable_model_schema method in the dify-official-plugins\models\xinference\models\tts\tts.py file did not return the model_properties. This is the code I fixed. ![Image](https://github.com/user-attachments/assets/bd004d21-526f-4f8b-90cd-9957f8e50a7c) It can now be displayed normally. ### ✔️ Error log _No response_
yindo added the bug label 2026-02-16 10:18:06 -05:00
yindo closed this issue 2026-02-16 10:18:06 -05:00
Author
Owner

@wxy8435 commented on GitHub (Mar 31, 2025):

这个bug,如果是本地docker 部署的 dify 镜像,应该怎么修正啊?

@wxy8435 commented on GitHub (Mar 31, 2025): 这个bug,如果是本地docker 部署的 dify 镜像,应该怎么修正啊?
Author
Owner

@937089773 commented on GitHub (Mar 31, 2025):

这个bug,如果是本地docker 部署的 dify 镜像,应该怎么修正啊?

这个有几种方式
1,可以将docker中的代码复制出来,docker cp,修改完成之后,将这个代码挂载到docker里面
2,如果不想那么复杂,可以在上面的操作中,将运行的容器保存为镜像,docker commit可以将容器保存为镜像
3,可以用源码的dockerfile构建镜像使用

@937089773 commented on GitHub (Mar 31, 2025): > 这个bug,如果是本地docker 部署的 dify 镜像,应该怎么修正啊? 这个有几种方式 1,可以将docker中的代码复制出来,docker cp,修改完成之后,将这个代码挂载到docker里面 2,如果不想那么复杂,可以在上面的操作中,将运行的容器保存为镜像,docker commit可以将容器保存为镜像 3,可以用源码的dockerfile构建镜像使用
Author
Owner

@scottywei commented on GitHub (Apr 8, 2025):

直接在docker/volumes/里面找到tts.py修改之后再启动,不生效呢?
After directly modifying tts.py in docker/volumes/ and then starting it again, the changes don’t take effect.

@scottywei commented on GitHub (Apr 8, 2025): 直接在docker/volumes/里面找到tts.py修改之后再启动,不生效呢? After directly modifying tts.py in docker/volumes/ and then starting it again, the changes don’t take effect.
Author
Owner

@sliqiao commented on GitHub (Apr 8, 2025):

官方能不能更新这个Bug,提供最新的Docker镜像,大家简单一点,这一行代码,直接把我们卡死了;
我们不想自己编译源码

@sliqiao commented on GitHub (Apr 8, 2025): 官方能不能更新这个Bug,提供最新的Docker镜像,大家简单一点,这一行代码,直接把我们卡死了; 我们不想自己编译源码
Author
Owner

@robin12jbj commented on GitHub (Jun 18, 2025):

Self Checks

  • This is only for bug report, if you would like to ask a question, please head to Discussions.[x] I have searched for existing issues Dify issues & Dify Official Plugins, including closed ones.[x] I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).[x] [FOR CHINESE USERS] 请务必使用英文提交 Issue,否则会被关闭。谢谢!:)[x] Please do not modify this template :) and fill in all the required fields.

Dify version

all

Plugin version

0.0.2

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

Image As shown in the above figure, after I selected the Cosyvoice model, I was unable to select the voice information. This is because the get_customizable_model_schema method in the dify-official-plugins\models\xinference\models\tts\tts.py file did not return the model_properties. This is the code I fixed.

Image It can now be displayed normally.

✔️ Error log

No response

excuse me, i run dify with docker, and xinference with conda env, how to use this project and edit the code? And where is the code you have editted? By the way, except dify sources code, how to use this project?

@robin12jbj commented on GitHub (Jun 18, 2025): > ### Self Checks > * [x] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general).[x] I have searched for existing issues [Dify issues](https://github.com/langgenius/dify/issues) & [Dify Official Plugins](https://github.com/langgenius/dify-official-plugins/issues), including closed ones.[x] I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)).[x] [FOR CHINESE USERS] 请务必使用英文提交 Issue,否则会被关闭。谢谢!:)[x] Please do not modify this template :) and fill in all the required fields. > > ### Dify version > all > > ### Plugin version > 0.0.2 > > ### Cloud or Self Hosted > Self Hosted (Docker) > > ### Steps to reproduce > ![Image](https://github.com/user-attachments/assets/13a20e14-1db9-4606-8fb9-3e377fd928bf) As shown in the above figure, after I selected the Cosyvoice model, I was unable to select the voice information. This is because the get_customizable_model_schema method in the dify-official-plugins\models\xinference\models\tts\tts.py file did not return the model_properties. This is the code I fixed. > > ![Image](https://github.com/user-attachments/assets/bd004d21-526f-4f8b-90cd-9957f8e50a7c) It can now be displayed normally. > > ### ✔️ Error log > _No response_ excuse me, i run dify with docker, and xinference with conda env, how to use this project and edit the code? And where is the code you have editted? By the way, except dify sources code, how to use this project?
Author
Owner

@robin12jbj commented on GitHub (Jun 18, 2025):

直接在docker/volumes/里面找到tts.py修改之后再启动,不生效呢? After directly modifying tts.py in docker/volumes/ and then starting it again, the changes don’t take effect.

have you solved this problem yet? are you tts.py in volumes\plugin_daemon\cwd\langgenius\xinference-0.0.3@.....\models\tts\tts.py?

@robin12jbj commented on GitHub (Jun 18, 2025): > 直接在docker/volumes/里面找到tts.py修改之后再启动,不生效呢? After directly modifying tts.py in docker/volumes/ and then starting it again, the changes don’t take effect. have you solved this problem yet? are you tts.py in volumes\plugin_daemon\cwd\langgenius\xinference-0.0.3@.....\models\tts\tts.py?
Author
Owner

@937089773 commented on GitHub (Jun 19, 2025):

Self Checks

  • This is only for bug report, if you would like to ask a question, please head to Discussions.[x] I have searched for existing issues Dify issues & Dify Official Plugins, including closed ones.[x] I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).[x] [FOR CHINESE USERS] 请务必使用英文提交 Issue,否则会被关闭。谢谢!:)[x] Please do not modify this template :) and fill in all the required fields.

Dify version

all

Plugin version

0.0.2

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

Image As shown in the above figure, after I selected the Cosyvoice model, I was unable to select the voice information. This is because the get_customizable_model_schema method in the dify-official-plugins\models\xinference\models\tts\tts.py file did not return the model_properties. This is the code I fixed.
Image It can now be displayed normally.

✔️ Error log

No response

excuse me, i run dify with docker, and xinference with conda env, how to use this project and edit the code? And where is the code you have editted? By the way, except dify sources code, how to use this project?

The environmental information I use is:
pycharm
conda -> python==3.12
The projects dify, Dify-official -plugins and Dify-plugins were downloaded respectively.
If you need to debug the project "dify-official-plugins", you need to install the go development environment
Some projects cannot run in windows. You will find this problem during the debugging process. It is best if they can run in the linux environment. The project can be run locally. You need to search for specific environmental information yourself on google.
You can deploy some projects through docker and then just debug a certain project.

@937089773 commented on GitHub (Jun 19, 2025): > > ### Self Checks > > > > > > * [x] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general).[x] I have searched for existing issues [Dify issues](https://github.com/langgenius/dify/issues) & [Dify Official Plugins](https://github.com/langgenius/dify-official-plugins/issues), including closed ones.[x] I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)).[x] [FOR CHINESE USERS] 请务必使用英文提交 Issue,否则会被关闭。谢谢!:)[x] Please do not modify this template :) and fill in all the required fields. > > ### Dify version > > all > > ### Plugin version > > 0.0.2 > > ### Cloud or Self Hosted > > Self Hosted (Docker) > > ### Steps to reproduce > > ![Image](https://github.com/user-attachments/assets/13a20e14-1db9-4606-8fb9-3e377fd928bf) As shown in the above figure, after I selected the Cosyvoice model, I was unable to select the voice information. This is because the get_customizable_model_schema method in the dify-official-plugins\models\xinference\models\tts\tts.py file did not return the model_properties. This is the code I fixed. > > ![Image](https://github.com/user-attachments/assets/bd004d21-526f-4f8b-90cd-9957f8e50a7c) It can now be displayed normally. > > ### ✔️ Error log > > _No response_ > > excuse me, i run dify with docker, and xinference with conda env, how to use this project and edit the code? And where is the code you have editted? By the way, except dify sources code, how to use this project? The environmental information I use is: pycharm conda -> python==3.12 The projects dify, Dify-official -plugins and Dify-plugins were downloaded respectively. If you need to debug the project "dify-official-plugins", you need to install the go development environment Some projects cannot run in windows. You will find this problem during the debugging process. It is best if they can run in the linux environment. The project can be run locally. You need to search for specific environmental information yourself on google. You can deploy some projects through docker and then just debug a certain project.
Author
Owner

@robin12jbj commented on GitHub (Jun 19, 2025):

Self Checks

  • This is only for bug report, if you would like to ask a question, please head to Discussions.[x] I have searched for existing issues Dify issues & Dify Official Plugins, including closed ones.[x] I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).[x] [FOR CHINESE USERS] 请务必使用英文提交 Issue,否则会被关闭。谢谢!:)[x] Please do not modify this template :) and fill in all the required fields.

Dify version

all

Plugin version

0.0.2

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

Image As shown in the above figure, after I selected the Cosyvoice model, I was unable to select the voice information. This is because the get_customizable_model_schema method in the dify-official-plugins\models\xinference\models\tts\tts.py file did not return the model_properties. This is the code I fixed.
Image It can now be displayed normally.

✔️ Error log

No response

excuse me, i run dify with docker, and xinference with conda env, how to use this project and edit the code? And where is the code you have editted? By the way, except dify sources code, how to use this project?

The environmental information I use is: pycharm conda -> python==3.12 The projects dify, Dify-official -plugins and Dify-plugins were downloaded respectively. If you need to debug the project "dify-official-plugins", you need to install the go development environment Some projects cannot run in windows. You will find this problem during the debugging process. It is best if they can run in the linux environment. The project can be run locally. You need to search for specific environmental information yourself on google. You can deploy some projects through docker and then just debug a certain project.

Image
I have already used dify in docker and it's linux environment. And where is the file I should modify and can I just modify and restart dify in docker to solve this problem?

@robin12jbj commented on GitHub (Jun 19, 2025): > > > ### Self Checks > > > > > > > > > * [x] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general).[x] I have searched for existing issues [Dify issues](https://github.com/langgenius/dify/issues) & [Dify Official Plugins](https://github.com/langgenius/dify-official-plugins/issues), including closed ones.[x] I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)).[x] [FOR CHINESE USERS] 请务必使用英文提交 Issue,否则会被关闭。谢谢!:)[x] Please do not modify this template :) and fill in all the required fields. > > > ### Dify version > > > all > > > ### Plugin version > > > 0.0.2 > > > ### Cloud or Self Hosted > > > Self Hosted (Docker) > > > ### Steps to reproduce > > > ![Image](https://github.com/user-attachments/assets/13a20e14-1db9-4606-8fb9-3e377fd928bf) As shown in the above figure, after I selected the Cosyvoice model, I was unable to select the voice information. This is because the get_customizable_model_schema method in the dify-official-plugins\models\xinference\models\tts\tts.py file did not return the model_properties. This is the code I fixed. > > > ![Image](https://github.com/user-attachments/assets/bd004d21-526f-4f8b-90cd-9957f8e50a7c) It can now be displayed normally. > > > ### ✔️ Error log > > > _No response_ > > > > > > excuse me, i run dify with docker, and xinference with conda env, how to use this project and edit the code? And where is the code you have editted? By the way, except dify sources code, how to use this project? > > The environmental information I use is: pycharm conda -> python==3.12 The projects dify, Dify-official -plugins and Dify-plugins were downloaded respectively. If you need to debug the project "dify-official-plugins", you need to install the go development environment Some projects cannot run in windows. You will find this problem during the debugging process. It is best if they can run in the linux environment. The project can be run locally. You need to search for specific environmental information yourself on google. You can deploy some projects through docker and then just debug a certain project. ![Image](https://github.com/user-attachments/assets/bff1b62c-0a19-4bf6-ae4e-fa8746fc2025) I have already used dify in docker and it's linux environment. And where is the file I should modify and can I just modify and restart dify in docker to solve this problem?
Author
Owner

@robin12jbj commented on GitHub (Jun 19, 2025):

Image
I run dify with docker. And above is the official volume mounted in docker. Is this tts.py is what you said above? If so, why can't I solve the problem by modify the tts.py and restart docker?

@robin12jbj commented on GitHub (Jun 19, 2025): ![Image](https://github.com/user-attachments/assets/3e1b51c2-0297-4b55-8eb0-467227026266) I run dify with docker. And above is the official volume mounted in docker. Is this tts.py is what you said above? If so, why can't I solve the problem by modify the tts.py and restart docker?
Author
Owner

@937089773 commented on GitHub (Jun 19, 2025):

Self Checks

  • This is only for bug report, if you would like to ask a question, please head to Discussions.[x] I have searched for existing issues Dify issues & Dify Official Plugins, including closed ones.[x] I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).[x] [FOR CHINESE USERS] 请务必使用英文提交 Issue,否则会被关闭。谢谢!:)[x] Please do not modify this template :) and fill in all the required fields.

Dify version

all

Plugin version

0.0.2

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

Image As shown in the above figure, after I selected the Cosyvoice model, I was unable to select the voice information. This is because the get_customizable_model_schema method in the dify-official-plugins\models\xinference\models\tts\tts.py file did not return the model_properties. This is the code I fixed.
Image It can now be displayed normally.

✔️ Error log

No response

excuse me, i run dify with docker, and xinference with conda env, how to use this project and edit the code? And where is the code you have editted? By the way, except dify sources code, how to use this project?

The environmental information I use is: pycharm conda -> python==3.12 The projects dify, Dify-official -plugins and Dify-plugins were downloaded respectively. If you need to debug the project "dify-official-plugins", you need to install the go development environment Some projects cannot run in windows. You will find this problem during the debugging process. It is best if they can run in the linux environment. The project can be run locally. You need to search for specific environmental information yourself on google. You can deploy some projects through docker and then just debug a certain project.

Image I have already used dify in docker and it's linux environment. And where is the file I should modify and can I just modify and restart dify in docker to solve this problem?

I wonder what your docker-compose.yaml configuration is like

Take my configuration as an example: The /app/storage configuration in the plugin_daemon container is mounted under d:/dify/plugin

So the TTS. Py in d: \ dify \ plugins \ CWD \ langgenius \ xinference - hundreds @ 04 dee9c8396689421df7537f73ebb4b26efbcb7e0b9c32b7f09236e84bb97b3 Under c\models\tts\tts.py.

It needs to be searched according to the specific version. Maybe yours is not 0.0.2, it might be 0.0.3

@937089773 commented on GitHub (Jun 19, 2025): > > > > ### Self Checks > > > > > > > > > > > > * [x] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general).[x] I have searched for existing issues [Dify issues](https://github.com/langgenius/dify/issues) & [Dify Official Plugins](https://github.com/langgenius/dify-official-plugins/issues), including closed ones.[x] I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)).[x] [FOR CHINESE USERS] 请务必使用英文提交 Issue,否则会被关闭。谢谢!:)[x] Please do not modify this template :) and fill in all the required fields. > > > > ### Dify version > > > > all > > > > ### Plugin version > > > > 0.0.2 > > > > ### Cloud or Self Hosted > > > > Self Hosted (Docker) > > > > ### Steps to reproduce > > > > ![Image](https://github.com/user-attachments/assets/13a20e14-1db9-4606-8fb9-3e377fd928bf) As shown in the above figure, after I selected the Cosyvoice model, I was unable to select the voice information. This is because the get_customizable_model_schema method in the dify-official-plugins\models\xinference\models\tts\tts.py file did not return the model_properties. This is the code I fixed. > > > > ![Image](https://github.com/user-attachments/assets/bd004d21-526f-4f8b-90cd-9957f8e50a7c) It can now be displayed normally. > > > > ### ✔️ Error log > > > > _No response_ > > > > > > > > > excuse me, i run dify with docker, and xinference with conda env, how to use this project and edit the code? And where is the code you have editted? By the way, except dify sources code, how to use this project? > > > > > > The environmental information I use is: pycharm conda -> python==3.12 The projects dify, Dify-official -plugins and Dify-plugins were downloaded respectively. If you need to debug the project "dify-official-plugins", you need to install the go development environment Some projects cannot run in windows. You will find this problem during the debugging process. It is best if they can run in the linux environment. The project can be run locally. You need to search for specific environmental information yourself on google. You can deploy some projects through docker and then just debug a certain project. > > ![Image](https://github.com/user-attachments/assets/bff1b62c-0a19-4bf6-ae4e-fa8746fc2025) I have already used dify in docker and it's linux environment. And where is the file I should modify and can I just modify and restart dify in docker to solve this problem? I wonder what your docker-compose.yaml configuration is like Take my configuration as an example: The /app/storage configuration in the plugin_daemon container is mounted under d:/dify/plugin So the TTS. Py in d: \ dify \ plugins \ CWD \ langgenius \ xinference - hundreds @ 04 dee9c8396689421df7537f73ebb4b26efbcb7e0b9c32b7f09236e84bb97b3 Under c\models\tts\tts.py. It needs to be searched according to the specific version. Maybe yours is not 0.0.2, it might be 0.0.3
Author
Owner

@937089773 commented on GitHub (Jun 19, 2025):

Image I run dify with docker. And above is the official volume mounted in docker. Is this tts.py is what you said above? If so, why can't I solve the problem by modify the tts.py and restart docker?

Yes, it can be done this way, but it should be noted that if the configuration in docker is changed and not synchronized to the host's disk, it will cause the docker container to be destroyed and the configuration to be lost. Of course, you can turn the modified container into an image, and this can also solve the problem

@937089773 commented on GitHub (Jun 19, 2025): > ![Image](https://github.com/user-attachments/assets/3e1b51c2-0297-4b55-8eb0-467227026266) I run dify with docker. And above is the official volume mounted in docker. Is this tts.py is what you said above? If so, why can't I solve the problem by modify the tts.py and restart docker? Yes, it can be done this way, but it should be noted that if the configuration in docker is changed and not synchronized to the host's disk, it will cause the docker container to be destroyed and the configuration to be lost. Of course, you can turn the modified container into an image, and this can also solve the problem
Author
Owner

@robin12jbj commented on GitHub (Jun 20, 2025):

Image

Image
I have changed the code as you say, but error in console. Is this normal?

@robin12jbj commented on GitHub (Jun 20, 2025): ![Image](https://github.com/user-attachments/assets/b1b0ff57-14dd-4836-bf21-1503fc824f98) ![Image](https://github.com/user-attachments/assets/44f1b8f6-5696-4adf-b3cf-76c236e589d2) I have changed the code as you say, but error in console. Is this normal?
Author
Owner

@dosubot[bot] commented on GitHub (Aug 21, 2025):

Hi, @937089773. I'm Dosu, and I'm helping the dify-official-plugins team manage their backlog and am marking this issue as stale.

Issue Summary:

  • You reported a bug in the xinference TTS plugin where voice options were not selectable after choosing the Cosyvoice model due to missing model_properties in get_customizable_model_schema.
  • You provided a code fix that correctly enabled voice info to display.
  • Other users discussed difficulties applying the fix in Docker environments, including volume mounting and container persistence.
  • You shared advice on debugging with conda and Docker, emphasizing the need to modify the correct tts.py file inside Docker volumes or commit container changes to keep fixes.
  • The issue was resolved with your code update and Docker usage guidance.

Next Steps:

  • Please confirm if this issue remains relevant with the latest version of the dify-official-plugins repository by commenting here.
  • If no further updates are provided, I will automatically close this issue in 5 days.

Thank you for your understanding and valuable contribution!

@dosubot[bot] commented on GitHub (Aug 21, 2025): Hi, @937089773. I'm [Dosu](https://dosu.dev), and I'm helping the dify-official-plugins team manage their backlog and am marking this issue as stale. **Issue Summary:** - You reported a bug in the xinference TTS plugin where voice options were not selectable after choosing the Cosyvoice model due to missing model_properties in get_customizable_model_schema. - You provided a code fix that correctly enabled voice info to display. - Other users discussed difficulties applying the fix in Docker environments, including volume mounting and container persistence. - You shared advice on debugging with conda and Docker, emphasizing the need to modify the correct tts.py file inside Docker volumes or commit container changes to keep fixes. - The issue was resolved with your code update and Docker usage guidance. **Next Steps:** - Please confirm if this issue remains relevant with the latest version of the dify-official-plugins repository by commenting here. - If no further updates are provided, I will automatically close this issue in 5 days. Thank you for your understanding and valuable contribution!
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langgenius/dify-official-plugins#111