Zao-chen
zzchat语音输入错误，输出好像是不是显存不够
确认在求助前尝试过以下办法（在[]内输入x确认）
- [x ] 我确认我使用的是最新的ZcChat
- [x ] 我已经认真阅读过README、WIKI以及教程视频
你遇到的问题 语音输入不了 聊天框显示error 下面有截图滴
描述你尝试过的方法 尝试问过ai软件但没什么用
附加上下文
[2025-03-15T07:09:57.699Z]
letta\agent.py:108: UserWarning: Tool rules only work reliably for the latest OpenAI models that support structured outputs.
[2025-03-15T07:09:58.016Z]
Letta.letta.agent - ERROR - step() failed messages = [Message(created_by_id=None, last_updated_by_id=None, created_at=datetime.datetime(2025, 3, 15, 7, 9, 57, 745275, tzinfo=datetime.timezone.utc), updated_at=None, id='message-d9407dbf-9cd2-455a-b0ca-8c2256f8bf9b', role=<MessageRole.user: 'user'>, content=[TextContent(type=<MessageContentType.text: 'text'>, text='{\n "type": "user_message",\n "message": "���",\n "time": "2025-03-15 03:09:57 PM �й���ʱ��+0800"\n}')], organization_id=None, agent_id='agent-4e4ddc63-366e-4229-81e7-6ed5a5e3ace3', model=None, name=None, tool_calls=None, tool_call_id=None, step_id=None, otid=None, tool_returns=None)] error = Request exceeds maximum context length (8399 > 8192 tokens)
[2025-03-15T07:09:58.048Z]
Letta.letta.agent - WARNING - context window exceeded with limit 8192, attempting to summarize (1/3
[2025-03-15T07:09:58.087Z]
Letta.letta.agent - INFO - System message token count=1277
[2025-03-15T07:09:58.087Z]
Letta.letta.agent - INFO - token_counts_no_system=[98, 93, 70, 67, 109, 91, 68, 119, 93, 68, 120, 92, 68, 137, 90, 68, 121, 92, 67, 113, 89, 68, 126, 94, 68, 126, 90, 68, 141, 91, 68, 133, 92, 68, 125, 88, 93, 148, 93, 74, 147, 92, 68, 135, 92, 74, 149, 95, 75, 145, 93, 81, 116, 94, 77, 167, 89, 77, 151, 90, 87, 166, 93, 76, 176, 90, 70, 120, 90, 76, 159, 93, 77, 162, 93, 67, 137, 91, 67, 135, 90, 67, 151, 96]
[2025-03-15T07:09:58.087Z]
Letta.letta.agent - INFO - desired_token_count_to_summarize=6790
[2025-03-15T07:09:58.087Z]
Letta.letta.agent - INFO - Evicting 71/85 messages... Letta.letta.agent - INFO - Attempting to summarize 71 messages of 85
[2025-03-15T07:09:58.804Z]
uvicorn.access - INFO - ::1:54561 - "GET /v1/health/ HTTP/1.1" 200 uvicorn.access - INFO - ::1:54561 - "GET /v1/health/ HTTP/1.1" 200
[2025-03-15T07:09:58.996Z]
uvicorn.access - INFO - ::1:54561 - "GET /v1/agents/?limit=1000 HTTP/1.1" 200 uvicorn.access - INFO - ::1:54561 - "GET /v1/agents/?limit=1000 HTTP/1.1" 200
[2025-03-15T07:09:59.747Z]
httpx - INFO - HTTP Request: POST https://inference.memgpt.ai/chat/completions "HTTP/1.1 500 Internal Server Error"
[2025-03-15T07:09:59.747Z]
Letta.letta.server.server - ERROR - Error in server._step: Internal Server Error
[2025-03-15T07:09:59.748Z]
Traceback (most recent call last): File "letta\server\server.py", line 365, in _step File "letta\tracing.py", line 188, in sync_wrapper File "letta\agent.py", line 744, in step File "letta\agent.py", line 991, in inner_step File "letta\agent.py", line 1089, in summarize_messages_inplace File "letta\memory.py", line 71, in summarize_messages File "letta\tracing.py", line 188, in sync_wrapper
[2025-03-15T07:09:59.748Z]
File "letta\llm_api\llm_api_tools.py", line 118, in wrapper File "letta\llm_api\llm_api_tools.py", line 61, in wrapper File "letta\llm_api\llm_api_tools.py", line 216, in create File "letta\llm_api\openai.py", line 445, in openai_chat_completions_request File "openai_utils_utils.py", line 279, in wrapper File "openai\resources\chat\completions\completions.py", line 914, in create File "openai_base_client.py", line 1242, in post File "openai_base_client.py", line 919, in request File "openai_base_client.py", line 1023, in _request openai.InternalServerError: Internal Server Error
[2025-03-15T07:09:59.749Z]
None Internal Server Error
[2025-03-15T07:09:59.749Z]
[2025-03-15T07:09:59.750Z]
Traceback (most recent call last): File "letta\server\server.py", line 1293, in send_message_to_agent File "asyncio\threads.py", line 25, in to_thread File "concurrent\futures\thread.py", line 59, in run File "letta\server\server.py", line 681, in send_messages File "letta\server\server.py", line 365, in _step File "letta\tracing.py", line 188, in sync_wrapper File "letta\agent.py", line 744, in step File "letta\agent.py", line 991, in inner_step File "letta\agent.py", line 1089, in summarize_messages_inplace File "letta\memory.py", line 71, in summarize_messages File "letta\tracing.py", line 188, in sync_wrapper File "letta\llm_api\llm_api_tools.py", line 118, in wrapper
[2025-03-15T07:09:59.750Z]
File "letta\llm_api\llm_api_tools.py", line 61, in wrapper File "letta\llm_api\llm_api_tools.py", line 216, in create File "letta\llm_api\openai.py", line 445, in openai_chat_completions_request File "openai_utils_utils.py", line 279, in wrapper File "openai\resources\chat\completions\completions.py", line 914, in create File "openai_base_client.py", line 1242, in post File "openai_base_client.py", line 919, in request File "openai_base_client.py", line 1023, in _request openai.InternalServerError: Internal Server Error 这个行吗，不知道日志文件在哪里
2025-03-15 15:23:10 [Info] [UnknownWindow] 获取到llmPost请求结果： ""
2025-03-15 15:23:10 [Info] [UnknownWindow] 接收到post信息： "<discarded>"
首先letta没有正常返回，你需要检查letta的设置。
2025-03-15 15:23:50 [Info] [galgamedialog] 使用whisper-asr-webservice语音识别……
2025-03-15 15:23:54 [Warning] [galgamedialog] 语音识别-whisper-错误: "Connection refused"
2025-03-15 15:23:54 [Warning] [galgamedialog] 语音识别-whisper-错误信息: ""
whisper-asr-webservice的配置也有问题，请查看whisper-asr-webservice的日志
刚才在找，打开了letta，没有找到设置（option）相关东西啊，whisper-asr-webservice根目录没有log的文件，是不是其他文件名
大佬我又新建了一个，现在打字交流没问题了，
还剩这个问题whisper-asr-webservice根目录没有log的文件，是不是其他文件名
你的whisper-asr-webservice部署成功了吗？用浏览器访问试试看。如果部署成功用浏览器访问会是一个文档地址 http://localhost:9000/
whisper-asr-webservice需要使用docker部署 docker安装指南
docker run -d --gpus all -p 9000:9000 \
-e ASR_MODEL=base \
-e ASR_ENGINE=openai_whisper \
onerahmet/openai-whisper-asr-webservice:latest-gpu
Docker's troubleshooting guide docker安装成功后命令行输入即可。
docker run -d --gpus all -p 9000:9000 \
-e ASR_MODEL=base \
-e ASR_ENGINE=openai_whisper \
onerahmet/openai-whisper-asr-webservice:latest-gpu
docker run -d --gpus all -p 9000:9000 -e ASR_MODEL=base -e ASR_ENGINE=openai_whisper onerahmet/openai-whisper-asr-webservice:latest-gpu
我已经尝试多次重装latta desktop 大佬现在是这样的，我尝试了删除 .latta 名下所有文件并关掉latta重新启动，发现可以进入了，但数据完全没有了，等重启计算机之后还是进不去latta，我的猜想是这样的 1 : .latta名下有上次的数据影响了第二次latta的打开（很可惜我不知道哪个文件代表这个数据，只能把.latta名下文件全删）不知道数据库文件名 当然我也尝试了删除temp名下的文件但好像用处不大 2：可能是我在退出latta时没有正确退出导致有些程序运行一般搁置对下次运行有影响（我的退出方法是关闭latta的窗口界面）因为桌面右下角没有看到latta的运行标志，没有正确退出latta可能是失败的原因
如果您看到了，希望能回复我