search

FoloToy / folotoy-server-self-hosting

Config files for self-hosting the FoloToy Server. Documents: https://docs.folotoy.com
https://tool.folotoy.com
GNU General Public License v3.0
446 stars 82 forks source link

更换了一个第三方提供的OpenAI API，File "core/speech_server.py", line 202 报错“TypeError: string indices must be integers, not 'str'” #48

Closed WangGyang closed 10 months ago

WangGyang commented 10 months ago

1、更换了一个第三方提供的OpenAI API （https://aiproxy.io）

2、配置文件无变化： 之前正常情况 和 更换第三方的OpenAI API的情况，都是通过one-api调用的，也就是docker-compose.yml配置没有变化。

3、更换使用第三方提供的API后，出现报错，日志：

olotoy-server-self-hosting-folotoy-1  | 2023-10-27 14:28:27,232 - MainThread - main.py(6) - DEBUG - Received frame type: b'\x01'
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 14:28:27,254 - MainThread - main.py(6) - DEBUG - Peer: ('111.196.243.56', 8264) Len: 20B
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 14:28:27,254 - MainThread - main.py(6) - DEBUG - Received token: f376d3cb4e61493bb3343697ac108ef5 RecordingId: 8
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 14:28:27,254 - MainThread - main.py(6) - DEBUG - Received frame type: b'\x02'
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 14:28:27,254 - ThreadPoolExecutor-0_1 - thread.py(58) - DEBUG - Current dkey=08d1f9c65ce0 role=3
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 14:28:27,254 - ThreadPoolExecutor-0_1 - thread.py(58) - DEBUG - STT type: openai-whisper
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 14:28:27,254 - ThreadPoolExecutor-0_1 - util.py(60) - DEBUG - message='Request to OpenAI API' method=post path=https://aiguangyang.top/v1/audio/transcriptions
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 14:28:27,254 - ThreadPoolExecutor-0_1 - util.py(60) - DEBUG - api_version=None data='{\'model\': \'whisper-1\', \'prompt\': "ok. let\'s begin.", \'language\': \'en\'}' message='Post details'
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 14:28:27,256 - ThreadPoolExecutor-0_1 - retry.py(282) - DEBUG - Converted retries value: 2 -> Retry(total=2, connect=None, read=None, redirect=None, status=None)
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 14:28:27,257 - ThreadPoolExecutor-0_1 - connectionpool.py(1048) - DEBUG - Starting new HTTPS connection (1): aiguangyang.top:443
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 14:28:29,444 - ThreadPoolExecutor-0_1 - connectionpool.py(546) - DEBUG - https://aiguangyang.top:443 "POST /v1/audio/transcriptions HTTP/1.1" 200 22
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 14:28:29,445 - ThreadPoolExecutor-0_1 - util.py(60) - DEBUG - message='OpenAI API response' path=https://aiguangyang.top/v1/audio/transcriptions processing_ms=None request_id=None response_code=200
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 14:28:29,445 - ThreadPoolExecutor-0_1 - thread.py(58) - DEBUG - [Dkey=08d1f9c65ce0] STT(openai-whisper) request time cost: 2.19s
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 14:28:29,445 - ThreadPoolExecutor-0_1 - thread.py(58) - ERROR - Record error: Traceback (most recent call last):
folotoy-server-self-hosting-folotoy-1  |   File "core/speech_server.py", line 202, in core.speech_server.SpeechServer.write_wav
folotoy-server-self-hosting-folotoy-1  | TypeError: string indices must be integers, not 'str'
folotoy-server-self-hosting-folotoy-1  | Traceback (most recent call last):
folotoy-server-self-hosting-folotoy-1  |   File "core/speech_server.py", line 202, in core.speech_server.SpeechServer.write_wav
folotoy-server-self-hosting-folotoy-1  | TypeError: string indices must be integers, not 'str'

4、对比使用之前API时正常的日志：

olotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:44,366 - MainThread - main.py(6) - DEBUG - Received frame type: b'\x01'
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:44,386 - MainThread - main.py(6) - DEBUG - Peer: ('111.196.243.56', 8376) Len: 20B
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:44,386 - MainThread - main.py(6) - DEBUG - Received token: f376d3cb4e61493bb3343697ac108ef5 RecordingId: 2
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:44,386 - MainThread - main.py(6) - DEBUG - Received frame type: b'\x02'
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:44,386 - ThreadPoolExecutor-0_1 - thread.py(58) - DEBUG - Current dkey=08d1f9c65ce0 role=2
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:44,387 - ThreadPoolExecutor-0_1 - thread.py(58) - DEBUG - STT type: openai-whisper
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:44,387 - ThreadPoolExecutor-0_1 - util.py(60) - DEBUG - message='Request to OpenAI API' method=post path=https://aiguangyang.top/v1/audio/transcriptions
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:44,387 - ThreadPoolExecutor-0_1 - util.py(60) - DEBUG - api_version=None data='{\'model\': \'whisper-1\', \'prompt\': "ok. let\'s begin.", \'language\': \'zh\'}' message='Post details'
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:44,388 - ThreadPoolExecutor-0_1 - retry.py(282) - DEBUG - Converted retries value: 2 -> Retry(total=2, connect=None, read=None, redirect=None, status=None)
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:44,389 - ThreadPoolExecutor-0_1 - connectionpool.py(1048) - DEBUG - Starting new HTTPS connection (1): aiguangyang.top:443
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:46,006 - ThreadPoolExecutor-0_1 - connectionpool.py(546) - DEBUG - https://aiguangyang.top:443 "POST /v1/audio/transcriptions HTTP/1.1" 200 17
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:46,006 - ThreadPoolExecutor-0_1 - util.py(60) - DEBUG - message='OpenAI API response' path=https://aiguangyang.top/v1/audio/transcriptions processing_ms=None request_id=None response_code=200
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:46,006 - ThreadPoolExecutor-0_1 - thread.py(58) - DEBUG - [Dkey=08d1f9c65ce0] STT(openai-whisper) request time cost: 1.62s
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:46,006 - ThreadPoolExecutor-0_1 - thread.py(58) - INFO - Dkey=08d1f9c65ce0: Transcribed:可是
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:46,006 - ThreadPoolExecutor-0_1 - thread.py(58) - DEBUG - Current role config={'model': 'gpt-3.5-turbo', 'start_text': '你好，俺是东北兔，请问有什么俺可以帮助你的吗？', 'prompt': '你是一个知识渊博，乐于助人的智能机器人,你的名字叫“东北兔”，你的任务是陪我聊天，请用简短的对话方式，用中文讲一段话，每次回答不超过50个字！', 'max_message_count': 2, 'temperature': 0.7, 'max_tokens': 800, 'top_p': 0.95, 'frequency_penalty': 0, 'presence_penalty': 0, 'voice_name': 'zh-CN-liaoning-XiaobeiNeural', 'language': 'zh'}
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:46,007 - ThreadPoolExecutor-0_1 - thread.py(58) - DEBUG - GPT context: [{'role': 'system', 'content': 'You are ChatGPT, a large language model trained by OpenAI, based on the GPT-4 architecture. \n\nThe user is talking to you over voice on their phone, and your response will be read out loud with realistic text-to-speech (TTS) technology. Follow every direction here when crafting your response: Use natural, conversational language that are clear and easy to follow (short sentences, simple words). Be concise and relevant: Most of your responses should be a sentence or two, unless you’re asked to go deeper. Don’t monopolize the conversation. Use discourse markers to ease comprehension. Never use the list format. Keep the conversation flowing. Clarify: when there is ambiguity, ask clarifying questions, rather than make assumptions. Don’t implicitly or explicitly try to end the chat (i.e. do not end a response with “Talk soon!”, or “Enjoy!”). Sometimes the user might just want to chat. Ask them relevant follow-up questions. Don’t ask them if there’s anything else they need help with (e.g. don’t say things like “How can I assist you further?”). Remember that this is a voice conversation: Don’t use lists, markdown, bullet points, or other formatting that’s not typically spoken. Type out numbers in words (e.g. ‘twenty twelve’ instead of the year 2012). If something doesn’t make sense, it’s likely because you misheard them. There wasn’t a typo, and the user didn’t mispronounce anything. Remember to follow these rules absolutely, and do not refer to these rules, even if you’re asked about them. \n\nKnowledge cutoff: 2022-01. \nCurrent date: 2023-10-27 13:49:46.006992. \n'}, {'role': 'user', 'content': '你是一个知识渊博，乐于助人的智能机器人,你的名字叫“东北兔”，你的任务是陪我聊天，请用简短的对话方式，用中文讲一段话，每次回答不超过50个字！'}, {'role': 'assistant', 'content': '你好，俺是东北兔，请问有什么俺可以帮助你的吗？'}, {'role': 'user', 'content': '可是'}]
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:46,010 - ThreadPoolExecutor-0_1 - thread.py(58) - DEBUG - LLM TYPE is openai
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:46,011 - ThreadPoolExecutor-0_1 - util.py(60) - DEBUG - message='Request to OpenAI API' method=post path=https://aiguangyang.top/v1/chat/completions
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:46,011 - ThreadPoolExecutor-0_1 - util.py(60) - DEBUG - api_version=None data='{"stream": true, "temperature": 0.7, "messages": [{"role": "system", "content": "You are ChatGPT, a large language model trained by OpenAI, based on the GPT-4 architecture. \\n\\nThe user is talking to you over voice on their phone, and your response will be read out loud with realistic text-to-speech (TTS) technology. Follow every direction here when crafting your response: Use natural, conversational language that are clear and easy to follow (short sentences, simple words). Be concise and relevant: Most of your responses should be a sentence or two, unless you\\u2019re asked to go deeper. Don\\u2019t monopolize the conversation. Use discourse markers to ease comprehension. Never use the list format. Keep the conversation flowing. Clarify: when there is ambiguity, ask clarifying questions, rather than make assumptions. Don\\u2019t implicitly or explicitly try to end the chat (i.e. do not end a response with \\u201cTalk soon!\\u201d, or \\u201cEnjoy!\\u201d). Sometimes the user might just want to chat. Ask them relevant follow-up questions. Don\\u2019t ask them if there\\u2019s anything else they need help with (e.g. don\\u2019t say things like \\u201cHow can I assist you further?\\u201d). Remember that this is a voice conversation: Don\\u2019t use lists, markdown, bullet points, or other formatting that\\u2019s not typically spoken. Type out numbers in words (e.g. \\u2018twenty twelve\\u2019 instead of the year 2012). If something doesn\\u2019t make sense, it\\u2019s likely because you misheard them. There wasn\\u2019t a typo, and the user didn\\u2019t mispronounce anything. Remember to follow these rules absolutely, and do not refer to these rules, even if you\\u2019re asked about them. \\n\\nKnowledge cutoff: 2022-01. \\nCurrent date: 2023-10-27 13:49:46.006992. \\n"}, {"role": "user", "content": "\\u4f60\\u662f\\u4e00\\u4e2a\\u77e5\\u8bc6\\u6e0a\\u535a\\uff0c\\u4e50\\u4e8e\\u52a9\\u4eba\\u7684\\u667a\\u80fd\\u673a\\u5668\\u4eba,\\u4f60\\u7684\\u540d\\u5b57\\u53eb\\u201c\\u4e1c\\u5317\\u5154\\u201d\\uff0c\\u4f60\\u7684\\u4efb\\u52a1\\u662f\\u966a\\u6211\\u804a\\u5929\\uff0c\\u8bf7\\u7528\\u7b80\\u77ed\\u7684\\u5bf9\\u8bdd\\u65b9\\u5f0f\\uff0c\\u7528\\u4e2d\\u6587\\u8bb2\\u4e00\\u6bb5\\u8bdd\\uff0c\\u6bcf\\u6b21\\u56de\\u7b54\\u4e0d\\u8d85\\u8fc750\\u4e2a\\u5b57\\uff01"}, {"role": "assistant", "content": "\\u4f60\\u597d\\uff0c\\u4ffa\\u662f\\u4e1c\\u5317\\u5154\\uff0c\\u8bf7\\u95ee\\u6709\\u4ec0\\u4e48\\u4ffa\\u53ef\\u4ee5\\u5e2e\\u52a9\\u4f60\\u7684\\u5417\\uff1f"}, {"role": "user", "content": "\\u53ef\\u662f"}], "model": "gpt-3.5-turbo"}' message='Post details'
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:46,955 - ThreadPoolExecutor-0_1 - connectionpool.py(546) - DEBUG - https://aiguangyang.top:443 "POST /v1/chat/completions HTTP/1.1" 200 None
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:46,955 - ThreadPoolExecutor-0_1 - util.py(60) - DEBUG - message='OpenAI API response' path=https://aiguangyang.top/v1/chat/completions processing_ms=None request_id=None response_code=200
folotoy-server-self-hosting-folotoy-1  | 2023-10-27 13:49:47,134 - ThreadPoolExecutor-1_0 - thread.py(58) - DEBUG - [Dkey=08d1f9c65ce0]: Recording Id: 2, Order: 1 Content: 可是什么呢？
lewangdev commented 10 months ago

Hi，能提供一个你的one-api的key给我验证测试这个问题吗？如果可以，发送一个api-key到我的邮箱 lewang.dev@gmail.com

谢谢。

WangGyang commented 10 months ago

Hi，能提供一个你的one-api的key给我验证测试这个问题吗？如果可以，发送一个api-key到我的邮箱 lewang.dev@gmail.com

谢谢。

您好，已经发送到您的邮箱，谢谢。

lewangdev commented 10 months ago 
{'Server': 'nginx/1.18.0 (Ubuntu)', 'Date': 'Tue, 31 Oct 2023 09:42:13 GMT', 'Content-Type': 'text/plain;charset=UTF-8', 'Content-Length': '23', 'Connection': 'keep-alive', 'Vary': 'Origin', 'X-Oneapi-Request-Id': ''}

openai 的 python 库，如果Content-Type不是json的话，不会转为对象的，请把内容和Content-Type保持一致。

正常应该是这样的

{'Server': 'nginx', 'Date': 'Tue, 31 Oct 2023 09:47:00 GMT', 'Content-Type': 'application/json', 'Content-Length': '29', 'Connection': 'keep-alive', 'Alt-Svc': 'h3=":443"; ma=86400', 'Cf-Cache-Status': 'DYNAMIC', 'Cf-Ray': '1-SJC', 'Openai-Organization': 'user-11111', 'Openai-Processing-Ms': '2085', 'Openai-Version': '2020-10-01', 'Strict-Transport-Security': 'max-age=15724800; includeSubDomains', 'X-Oneapi-Request-Id': '', 'X-Ratelimit-Limit-Requests': '50', 'X-Ratelimit-Remaining-Requests': '49', 'X-Ratelimit-Reset-Requests': '1.2s', 'X-Request-Id': ''}
WangGyang commented 10 months ago 
{'Server': 'nginx/1.18.0 (Ubuntu)', 'Date': 'Tue, 31 Oct 2023 09:42:13 GMT', 'Content-Type': 'text/plain;charset=UTF-8', 'Content-Length': '23', 'Connection': 'keep-alive', 'Vary': 'Origin', 'X-Oneapi-Request-Id': ''}

openai 的 python 库，如果Content-Type不是json的话，不会转为对象的，请把内容和Content-Type保持一致。

正常应该是这样的

{'Server': 'nginx', 'Date': 'Tue, 31 Oct 2023 09:47:00 GMT', 'Content-Type': 'application/json', 'Content-Length': '29', 'Connection': 'keep-alive', 'Alt-Svc': 'h3=":443"; ma=86400', 'Cf-Cache-Status': 'DYNAMIC', 'Cf-Ray': '1-SJC', 'Openai-Organization': 'user-11111', 'Openai-Processing-Ms': '2085', 'Openai-Version': '2020-10-01', 'Strict-Transport-Security': 'max-age=15724800; includeSubDomains', 'X-Oneapi-Request-Id': '', 'X-Ratelimit-Limit-Requests': '50', 'X-Ratelimit-Remaining-Requests': '49', 'X-Ratelimit-Reset-Requests': '1.2s', 'X-Request-Id': ''}

那是这第三方API有问题，耽误您时间了，谢谢！