What causes "Aliyun音频服务异常，请检查ALIYUN_TOKEN和ALIYUN_APPKEY是否过期。" in binary-husky/gpt_academic?

Generate a fresh Aliyun NLS token and put it in ALIYUN_TOKEN, or leave it empty and configure valid ALIYUN_ACCESSKEY and ALIYUN_SECRET so get_token() can create one.

binary-husky/gpt_academic · error · RuntimeError

Aliyun音频服务异常，请检查ALIYUN_TOKEN和ALIYUN_APPKEY是否过期。

Error message

Aliyun音频服务异常，请检查ALIYUN_TOKEN和ALIYUN_APPKEY是否过期。

What it means

The audio plugin exits its capture loop and re-raises self.stop_msg at crazy_functions/Audio_Assistant.py:166. For this message, AliyunASR set stop_msg in crazy_functions/live_audio/aliyunASR.py:218-220 after the Aliyun NLS websocket transcriber closed. It is a generic ASR session failure, but this implementation labels every close as an expired ALIYUN_TOKEN or ALIYUN_APPKEY problem.

Source

Thrown at crazy_functions/Audio_Assistant.py:166

                self.commit_wd.begin_watch()
                chatbot[-1] = list(chatbot[-1])
                chatbot[-1] = [self.buffered_sentence, "[ 等待GPT响应 ]"]
                yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
                # add gpt task 创建子线程请求gpt，避免线程阻塞
                history = chatbot2history(chatbot)
                self.agt.add_async_gpt_task(self.buffered_sentence, len(chatbot)-1, llm_kwargs, history, system_prompt)

                self.buffered_sentence = ""
                chatbot.append(["[ 请讲话 ]", "[ 正在等您说完问题 ]"])
                yield from update_ui(chatbot=chatbot, history=history) # 刷新界面

            if not self.event_on_result_chg.is_set() and not self.event_on_entence_end.is_set() and not self.event_on_commit_question.is_set():
                visualize_audio(chatbot, self.audio_shape)
                yield from update_ui(chatbot=chatbot, history=history) # 刷新界面

        if len(self.stop_msg) != 0:
            raise RuntimeError(self.stop_msg)



@CatchException
def Audio_Assistant(txt, llm_kwargs, plugin_kwargs, chatbot, history, system_prompt, user_request):
    # pip install -U openai-whisper
    chatbot.append(["对话助手函数插件：使用时，双手离开鼠标键盘吧", "音频助手, 正在听您讲话（点击“停止”键可终止程序）..."])
    yield from update_ui(chatbot=chatbot, history=history) # 刷新界面

    # 尝试导入依赖，如果缺少依赖，则给出安装建议
    try:
        import nls
        from scipy import io
    except:
        chatbot.append(["导入依赖失败", "使用该模块需要额外依赖, 安装方法:```pip install --upgrade aliyun-python-sdk-core==2.13.3 pyOpenSSL webrtcvad scipy git+https://github.com/aliyun/alibabacloud-nls-python-sdk.git```"])
        yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
        return

View on GitHub (pinned to d6bde0fa54)

Solutions

Generate a fresh Aliyun NLS token and put it in ALIYUN_TOKEN, or leave it empty and configure valid ALIYUN_ACCESSKEY and ALIYUN_SECRET so get_token() can create one.
Confirm ALIYUN_APPKEY is the appkey of an enabled Intelligent Speech Interaction project and belongs to the same account/token.
Check network access to wss://nls-gateway.aliyuncs.com/ws/v1 and temporarily disable proxies that cannot tunnel websockets.
Reinstall the required Aliyun SDK with the command shown by the plugin, then reproduce the close in a minimal NlsSpeechTranscriber script.
Improve on_error/on_close handling so the actual Aliyun error code is stored instead of always reporting credential expiration.

Example fix

# before
def test_on_close(self, *args):
    self.aliyun_service_ok = False

def test_on_error(self, message, *args):
    logging.error("on_error args=>{}".format(args))

# after
def test_on_error(self, message, *args):
    try:
        detail = json.loads(message)
        self.stop_msg = "Aliyun ASR error: " + str(detail)
    except Exception:
        self.stop_msg = "Aliyun ASR error: " + str(message)
    self.aliyun_service_ok = False

Defensive patterns

Strategy: try-catch

Validate before calling

TOKEN, APPKEY = get_conf("ALIYUN_TOKEN", "ALIYUN_APPKEY")
if not APPKEY:
    raise ValueError("ALIYUN_APPKEY is required")
if not TOKEN and not all(get_conf("ALIYUN_ACCESSKEY", "ALIYUN_SECRET")):
    raise ValueError("Provide ALIYUN_TOKEN or valid ALIYUN_ACCESSKEY/ALIYUN_SECRET")

Try / catch

try:
    yield from ia.begin(llm_kwargs, plugin_kwargs, chatbot, history, system_prompt)
except RuntimeError as e:
    if "Aliyun音频服务异常" in str(e):
        chatbot.append(["音频助手停止", "请更新阿里云 NLS Token/AppKey 后重试。"])
        yield from update_ui(chatbot=chatbot, history=history)
    else:
        raise

Prevention

Refresh Aliyun NLS tokens before long voice sessions because they expire.
Keep the appkey and token in the same NLS project.
Test websocket connectivity to nls-gateway.aliyuncs.com before starting a demo.
Capture on_error payloads so session failures are diagnosable.

When it happens

Trigger: Starting or streaming through nls.NlsSpeechTranscriber against wss://nls-gateway.aliyuncs.com/ws/v1 fails or the websocket calls on_close. Typical causes are an expired NLS token, a token that does not match ALIYUN_APPKEY, an appkey that is not enabled for the project, or a blocked/unstable WSS connection.

Common situations: ALIYUN_TOKEN was generated earlier and has expired; ALIYUN_TOKEN is empty and CreateToken failed because ALIYUN_ACCESSKEY/ALIYUN_SECRET are wrong; the appkey belongs to a different NLS project; corporate proxy or firewall blocks websocket traffic; the SDK dependency from the Aliyun git repository is missing or incompatible.

Related errors

AI-assisted analysis of binary-husky/gpt_academic@d6bde0fa54 (2026-08-14). Data as JSON: /api/errors/26ac1b78fda7dd2e. Report an issue: GitHub.