BabelDuck
BabelDuck copied to clipboard
Orenoid
docker部署后快捷指令问题
我用docker部署之后，快捷指令似乎无法使用，语音输入也是一样，使用的siliconflow的deepseek模型，这个功能是只有openai才能使用吗？
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
Title: issue with shortcut command after docker deployment
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
Is this feature only openai used in deepseek model of siliconflow?
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
Is this a bug if the page access does not have a logo after docker is deployed locally?
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
Is this a bug if the page access does not have a logo after docker local deployment?
Not only this, it is not only possible to modify and translate AI, but also to record.
快捷指令和语音输入不能工作可能是环境变量配置有错误，麻烦截图看一下 .env 配置（API KEY 等信息请打码，主要是确认下是否有遗漏或者错误格式）
P.S. 这其实算是个遗留问题，未来解决后，原则上 .env 是不需要配置什么 AI 相关配置的，我这段时间实在是比较忙，还没空去解决这个问题
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
Shortcut commands and voice input cannot work, which may be because the environment variable configuration is incorrect. Please take a screenshot and look at the .env configuration (please code the API KEY and other information, mainly to confirm whether there are any omissions or incorrect formats)
我也碰到类似问题，macbook+docker部署，env 配置了stt的API [SK，用语音输入会显示“undefine"，请帮忙看看，谢谢。
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
I also encountered a similar problem, macbook+docker deployment, env configured stt's API [SK, and the voice input will display "undefined". Please help, thank you.
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
Looking forward to the author's update. The project is very good.
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
Is this a bug if the page access does not have a logo after docker local deployment?
I've reappeared here, it's indeed a bug, to be checked
快捷指令和语音输入不能工作可能是环境变量配置有错误，麻烦截图看一下 .env 配置（API KEY 等信息请打码，主要是确认下是否有遗漏或者错误格式）
P.S. 这其实算是个遗留问题，未来解决后，原则上 .env 是不需要配置什么 AI 相关配置的，我这段时间实在是比较忙，还没空去解决这个问题
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
Shortcut commands and voice input cannot work, which may be because the environment variable configuration is incorrect. Please take a screenshot and look at the .env configuration (please code the API KEY and other information, mainly to confirm whether there are any omissions or incorrect formats)
P.S. This is actually a legacy problem. After solving it in the future, in principle, .env does not need to configure any AI-related configurations. I am really busy during this period and have not had time to solve this problem.
@dacktang 快捷指令功能目前需要把你上面空着的那几个环境变量填上才能使用（遗留问题，待解决），供参考： OPENAI_CHAT_COMPLETION_URL=https://api.siliconflow.cn/v1/chat/completions OPENAI_API_KEY=xxxxxx OPENAI_MODEL_NAME=deepseek-ai/DeepSeek-V2.5
这里的配置不一定得用 OpenAI 的服务，只要是兼容 OpenAI chat 接口格式的平台就行，比如我个人使用的就是 siliconflow，另外这里选择的模型需要支持 json mode
至于语音转文本的问题，我之后排查下跟上面另一位朋友提到的返回 undefined 的现象是否同一个问题，再统一回复。 另外我这边正在着手开发集成浏览器本地 STT 的功能，预计会在一两周内上线，到时就不需要依赖 siliconflow 的语音转文本接口了。
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
The @dacktang shortcut function currently needs to fill in the several environment variables that are empty above you before using (legacy issues, to be solved), for reference: OPENAI_CHAT_COMPLETION_URL=https://api.siliconflow.cn/v1/chat/completes OPENAI_API_KEY=xxxxxx OPENAI_MODEL_NAME=deepseek-ai/DeepSeek-V2.5
The configuration here does not necessarily require OpenAI services. As long as it is a platform that is compatible with the OpenAI chat interface format, for example, I personally use siliconflow. In addition, the model here needs to support json mode
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
@Caicai-cailo @dacktang logo The problem of not loading has been fixed. Just pull it or build the latest image by yourself.
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
@Caicai-cailo @dacktang The problem of the logo cannot be loaded has been fixed. Just pull it or build the latest image by yourself.
OK, thanks
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
@Caicai-cailo @dacktang The problem of the logo cannot be loaded has been fixed. Just pull it or build the latest image by yourself.
Thanks.
语音无法识别，docker日志里报错
⨯ Error: 'sharp' is required to be installed in standalone mode for the image optimization to function correctly. Read more at: https://nextjs.org/docs/messages/sharp-missing-in-production
(node:1) ExperimentalWarning: buffer.File is an experimental feature and might change at any time
(Use node --trace-warnings ... to show where the warning was created)
⨯ Error: 'sharp' is required to be installed in standalone mode for the image optimization to function correctly. Read more at: https://nextjs.org/docs/messages/sharp-missing-in-production
⨯ Error: 'sharp' is required to be installed in standalone mode for the image optimization to function correctly. Read more at: https://nextjs.org/docs/messages/sharp-missing-in-production
⨯ Error: 'sharp' is required to be installed in standalone mode for the image optimization to function correctly. Read more at: https://nextjs.org/docs/messages/sharp-missing-in-production
⨯ Error: 'sharp' is required to be installed in standalone mode for the image optimization to function correctly. Read more at: https://nextjs.org/docs/messages/sharp-missing-in-production
我也遇到了,语音无法识别,识别后返回undefined 并且docker logs报错信息 ⨯ Error: 'sharp' is required to be installed in standalone mode for the image optimization to function correctly. Read more at: https://nextjs.org/docs/messages/sharp-missing-in-production
看起来是 SicliconFlow 那边把语音识别模型给下架了，接下来会接入浏览器本地 STT 和 Azure STT，移除原有的语音识别接口。 在此之前建议先用各个系统自带的语音输入或输入法提供的语音输入过渡。
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
It seems that SicliconFlow has removed the voice recognition model from the shelves. Next, it will connect to the browser's local STT and Azure STT to remove the original voice recognition interface. Before this, it is recommended to use the voice input provided by the voice input method provided by each system or the voice input transition.
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
My interface using OPENAI will show "Chat model service is temporarily unavailable, please try again later". Is there any problem with where to set it up?