在Google Cloud Run运行微软TTS遇RuntimeError及连接问题求助

云上老周

2026-7-18

问题描述

在Google Cloud Run上运行微软文本转语音（TTS）服务时出现两类异常，仅通过Cloud Build构建部署时触发，本地构建镜像直接部署则完全正常：

初始错误：平台初始化失败

报错栈：

Traceback (most recent call last):
  File "/code/app/./speech/backend.py", line 42, in save_text_to_speech
    speech_api.speech()
  File "/code/app/./speech/speech_api.py", line 266, in speech
    synthesizer = SpeechSynthesizer(speech_config=self.speech_config, audio_config=audio_config)
  File "/usr/local/lib/python3.9/site-packages/azure/cognitiveservices/speech/speech.py", line 1598, in __init__
    self._impl = self._get_impl(impl.SpeechSynthesizer, speech_config, audio_config,
  File "/usr/local/lib/python3.9/site-packages/azure/cognitiveservices/speech/speech.py", line 1703, in _get_impl
    _impl = synth_type._from_config(speech_config._impl, None if audio_config is None else audio_config._impl)
RuntimeError: Runtime error: Failed to initialize platform (azure-c-shared). Error: 2153

触发代码片段：

audio_config = AudioOutputConfig(filename=file_name)
synthesizer = SpeechSynthesizer(speech_config=self.speech_config, audio_config=audio_config)
synthesizer.speak_ssml(self.input_data['text'])

本地环境：

MacOS 11.6.2
Docker v20.10.10
已尝试操作：清除kaniko缓存、切换Cloud Build构建工具为gcr.io/cloud-builders/docker，问题未解决。

2023-07-18更新：WebSocket连接错误

更新后的Dockerfile：

FROM python:3.11

WORKDIR /app


RUN apt-get update && \
    apt-get install -y build-essential libssl-dev ca-certificates libasound2 wget && \
    wget -O - https://www.openssl.org/source/openssl-1.1.1u.tar.gz | tar zxf - && \
    cd openssl-1.1.1u && \
    ./config --prefix=/usr/local && \
    make -j $(nproc) && \
    make install_sw install_ssldirs && \
    ldconfig -v && \
    export SSL_CERT_DIR=/etc/ssl/certs && \
    cd ../ && \
    rm -rf openssl-1.1.1u && \
    pip install --no-cache-dir azure-cognitiveservices-speech==1.30.0
COPY . /app
CMD ["python3", "app.py"]

测试代码（基于官方示例）：

import os
import azure.cognitiveservices.speech as speechsdk

speech_config = speechsdk.SpeechConfig(subscription=os.environ.get('SPEECH_KEY'), region=os.environ.get('SPEECH_REGION'))
audio_config = speechsdk.audio.AudioOutputConfig(use_default_speaker=True)
print('key:'+os.environ.get('SPEECH_KEY'))
print('region:'+os.environ.get('SPEECH_REGION'))
speech_config.speech_synthesis_voice_name='en-US-JennyNeural'

speech_synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config, audio_config=audio_config)

text = "Hello world!"
speech_synthesis_result = speech_synthesizer.speak_text_async(text).get()

if speech_synthesis_result.reason == speechsdk.ResultReason.SynthesizingAudioCompleted:
    print("Speech synthesized for text [{}]".format(text))
elif speech_synthesis_result.reason == speechsdk.ResultReason.Canceled:
    cancellation_details = speech_synthesis_result.cancellation_details
    print("Speech synthesis canceled: {}".format(cancellation_details.reason))
    if cancellation_details.reason == speechsdk.CancellationReason.Error:
        if cancellation_details.error_details:
            print("Error details: {}".format(cancellation_details.error_details))
            print("Did you set the speech resource key and region values?")

新报错信息：

Speech synthesis canceled: CancellationReason.Error
Error details: Connection failed (no connection to the remote host). Internal error: 1. Error details: Failed with error: WS_OPEN_ERROR_UNDERLYING_IO_OPEN_FAILED
wss://southeastasia.tts.speech.microsoft.com/cognitiveservices/websocket/v1
X-ConnectionId: c4955d953f8e480c906061e6219eb8fd USP state: Sending. Received audio size: 0 bytes.
Did you set the speech resource key and region values?

已确认SPEECH_KEY和SPEECH_REGION配置正确（控制台已打印验证）。

解决方案

针对平台初始化错误（Error 2153）

修复系统依赖缺失
Azure Speech SDK的azure-c-shared组件依赖特定系统库，Cloud Build环境可能存在依赖遗漏。修改Dockerfile，安装完整依赖并避免手动编译OpenSSL（手动编译易导致链接路径不兼容）：

RUN apt-get update && apt-get install -y --no-install-recommends \
    libssl-dev \
    libasound2 \
    ca-certificates \
    && rm -rf /var/lib/apt/lists/*

指定库加载路径
在Dockerfile中添加环境变量，强制SDK使用系统标准库路径：

ENV LD_LIBRARY_PATH=/usr/lib/x86_64-linux-gnu:$LD_LIBRARY_PATH

针对WebSocket连接错误（WS_OPEN_ERROR_UNDERLYING_IO_OPEN_FAILED）

强制使用REST协议
Cloud Run环境对WebSocket出站连接可能有限制，修改代码禁用WebSocket，改用REST API：

speech_config.set_property(speechsdk.PropertyId.SpeechServiceConnection_EnableWebSocket, "false")

适配Cloud Run无音频设备环境
Cloud Run容器没有默认音频输出设备，将音频输出改为文件：

audio_config = speechsdk.audio.AudioOutputConfig(filename="output.wav")

验证网络与证书

若使用VPC连接器，确保其允许访问微软TTS服务域名（*.tts.speech.microsoft.com）
在Dockerfile中添加证书更新步骤：
```
RUN update-ca-certificates -f
```

统一优化后的Dockerfile

FROM python:3.11-slim

WORKDIR /app

# 安装系统依赖并更新证书
RUN apt-get update && apt-get install -y --no-install-recommends \
    build-essential \
    libssl-dev \
    libasound2 \
    ca-certificates \
    && update-ca-certificates -f \
    && rm -rf /var/lib/apt/lists/*

# 设置环境变量
ENV LD_LIBRARY_PATH=/usr/lib/x86_64-linux-gnu:$LD_LIBRARY_PATH
ENV SSL_CERT_DIR=/etc/ssl/certs

# 安装Python依赖
RUN pip install --no-cache-dir azure-cognitiveservices-speech==1.30.0

COPY . /app

CMD ["python3", "app.py"]

优化后的测试代码

import os
import azure.cognitiveservices.speech as speechsdk

speech_config = speechsdk.SpeechConfig(subscription=os.environ.get('SPEECH_KEY'), region=os.environ.get('SPEECH_REGION'))
# 禁用WebSocket，改用REST协议
speech_config.set_property(speechsdk.PropertyId.SpeechServiceConnection_EnableWebSocket, "false")
# 输出到文件（适配Cloud Run无音频设备环境）
audio_config = speechsdk.audio.AudioOutputConfig(filename="output.wav")

speech_config.speech_synthesis_voice_name='en-US-JennyNeural'

speech_synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config, audio_config=audio_config)

text = "Hello world!"
speech_synthesis_result = speech_synthesizer.speak_text_async(text).get()

if speech_synthesis_result.reason == speechsdk.ResultReason.SynthesizingAudioCompleted:
    print(f"Speech synthesized for text [{text}]")
elif speech_synthesis_result.reason == speechsdk.ResultReason.Canceled:
    cancellation_details = speech_synthesis_result.cancellation_details
    print(f"Speech synthesis canceled: {cancellation_details.reason}")
    if cancellation_details.reason == speechsdk.CancellationReason.Error:
        if cancellation_details.error_details:
            print(f"Error details: {cancellation_details.error_details}")

内容的提问来源于stack exchange，提问作者Nori