Windows下使用Azure Speech SDK如何用Python获取麦克风设备ID创建AudioConfig

你可以通过pycaw库调用Windows原生音频API获取符合Azure Speech SDK要求的麦克风设备ID，无需依赖C++/C#组件。

前置依赖安装 #

执行以下命令安装所需的第三方库：

pip install pycaw comtypes

获取音频输入设备列表及对应ID的代码 #

import comtypes
from pycaw.pycaw import AudioUtilities, IMMDeviceEnumerator, EDataFlow, DEVICE_STATE
from comtypes import CoInitialize, CoUninitialize

def get_all_capture_devices():
    CoInitialize()
    device_list = []
    try:
        # 初始化设备枚举器
        enumerator = comtypes.CoCreateInstance(
            IMMDeviceEnumerator._iid_,
            interface=IMMDeviceEnumerator,
            clsctx=comtypes.CLSCTX_ALL
        )
        # 枚举所有处于激活状态的音频输入设备
        endpoint_collection = enumerator.EnumAudioEndpoints(
            EDataFlow.eCapture, 
            DEVICE_STATE.ACTIVE
        )
        device_count = endpoint_collection.GetCount()
        for i in range(device_count):
            dev = endpoint_collection.Item(i)
            dev_id = dev.GetId()
            dev_name = AudioUtilities.GetDeviceName(dev)
            device_list.append({
                "device_name": dev_name,
                "azure_device_id": dev_id
            })
    finally:
        CoUninitialize()
    return device_list

# 调用示例：打印所有可用麦克风的名称和对应ID
if __name__ == "__main__":
    all_mics = get_all_capture_devices()
    for mic in all_mics:
        print(f"麦克风名称：{mic['device_name']}")
        print(f"Azure SDK可用device_id：{mic['azure_device_id']}\n")

上述代码返回的azure_device_id无需额外格式化，可直接作为参数传入AudioConfig的device_name字段

调用Azure Speech SDK示例 #

拿到目标设备ID后，按以下方式初始化音频配置即可：

import azure.cognitiveservices.speech as speechsdk

# 替换为你的语音服务配置
speech_config = speechsdk.SpeechConfig(
    subscription="YOUR_AZURE_SPEECH_KEY",
    region="YOUR_SERVICE_REGION"
)
# 替换为上面获取到的目标麦克风的ID
audio_config = speechsdk.audio.AudioConfig(
    device_name="YOUR_TARGET_MIC_DEVICE_ID"
)

# 后续正常初始化语音翻译客户端即可

内容的提问来源于stack exchange，提问作者Karthik Sunil