功能概述
允许用户选择通过麦克风（microphone）或者获取系统声音（system_sound）作为语音输入源。
对输入的语音进行识别，得到对应的文本内容。
将识别出的文本翻译成指定的目标语言。
把翻译后的文本转换为语音并播放出来。

import speech_recognition as sr
from gtts import gTTS
from googletrans import Translator
import os
import time
import pyaudio
import pyshark# 创建识别器对象
r = sr.Recognizer()
# 创建翻译器对象
translator = Translator()def recognize_speech(language, input_type="microphone"):if input_type == "microphone":with sr.Microphone() as source:print(f"请说话（{language}）：")audio = r.listen(source)elif input_type == "system_sound":CHUNK = 1024FORMAT = pyaudio.paInt16CHANNELS = 1RATE = 44100p = pyaudio.PyAudio()stream = p.open(format=FORMAT,channels=CHANNELS,rate=RATE,input=True,frames_per_buffer=CHUNK)print(f"正在获取系统声音（{language}），请在系统中播放相关音频...")frames = []for i in range(0, int(RATE / CHUNK * 5)):  # 获取5秒钟的音频数据，可根据需要调整data = stream.read(CHUNK)frames.append(data)stream.stop_stream()stream.close()p.terminate()audio_data = sr.AudioData(b''.join(frames), RATE, FORMAT)else:raise ValueError("无效的输入类型，可选'microphone'或'system_sound'")try:if input_type == "microphone":text = r.recognize_google(audio, language=language)elif input_type == "system_sound":text = r.recognize_google(audio_data, language=language)print(f"识别结果： {text}")return textexcept sr.UnknownError:print("无法识别语音")except sr.RequestError as e:print(f"请求错误：{e}")def translate_text(text, target_language):translation = translator.translate(text, dest=target_language)print(f"翻译结果： {translation.text}")return translation.textdef speak_text(text, language):tts = gTTS(text=text, lang=language)tts.save("output.mp3")os.system("mpg123 output.mp3")os.remove("output.mp3")if __name__ == "__main__":source_language = "en-US"  # 源语言，这里设置为英语（美国），可根据需要修改target_language = "zh-CN"  # 目标语言，这里设置为中文（中国大陆），可根据需要修改input_option = input("请选择输入方式（microphone/system_sound）：")while True:recognized_text = recognize_speech(source_language, input_type=input_option)if recognized_text:translated_text = translate_text(recognized_text, target_language)speak_text(translated_text, target_language)time.sleep(2)  # 暂停2秒，避免过于频繁的识别