基于pyaudio实现的录音功能

提示：文章写完后，目录可以自动生成，如何生成可参考右边的帮助文档

前言

最近由于有实现录音和播放功能的需求，最近将pyaudio实现录音和播音功能进行实现，特此进行记录

提示：以下是本篇文章正文内容，下面案例可供参考

二、详细代码

代码如下（示例）：

import threading
import time
import pyaudio
from pyaudio import Stream
import wave
from typing import Optional


class SoundRecord:
    """
    功能:实现录音功能
    """

    def __init__(self):
        self.__output_file = None  # 输出文件地址
        self.__chunk = 1024  # 读取数据段
        self.__sample_format = pyaudio.paInt16  # 采样格式
        self.__channels = 1  # 声道数
        self.__sample_rate = 44100  # 采样率44.1kz
        self.__input_stream: Optional[Stream] = None  # 输入音频流
        self.__output_stream: Optional[Stream] = None  # 输出音频流
        self.__input_audio = None
        self.__output_audio = None
        self.__record_state = True
        self.__play_flag = True
        self.__frame = list()
        self.__wf = None

    def open_audio_input_stream(self):
        """
        功能:打开录音stream音频流
        """
        # 创建Pyaudio对象
        self.__input_audio = pyaudio.PyAudio()
        # 打开音频流
        try:
            self.__input_stream = self.__input_audio.open(format=self.__sample_format,
                                                          channels=self.__channels,
                                                          rate=self.__sample_rate,
                                                          frames_per_buffer=self.__chunk,
                                                          input=True)
            return True
        except OSError:  # 没有找到设备
            return False

    def open_audio_output_stream(self, file_path):
        """
        功能:打开播放stream音频流
        """
        self.__wf = wave.open(file_path, 'rb')
        self.__output_audio = pyaudio.PyAudio()
        try:
            self.__output_stream = self.__output_audio.open(format=self.__output_audio.get_format_from_width(self.__wf.getsampwidth()),
                                                            channels=self.__wf.getnchannels(),
                                                            rate=self.__wf.getframerate(),
                                                            output=True)
            return True

        except OSError:  # 没有找到设备
            return False

    def play_file_thread(self):
        """
        功能：播放文件线程
        """
        play_thread = threading.Thread(target=self.play_file)
        play_thread.start()

    def play_file(self):
        """
        功能:播放文件
        """
        data = self.__wf.readframes(1024)  # 将文件写入进播放音频流
        while len(data) > 0:
            self.__output_stream.write(data)
            data = self.__wf.readframes(1024)
        self.__output_stream.start_stream()
        self.__play_flag = False

    def close_output_thread(self):
        """
        功能：释放相关资源线程
        """
        output_thread = threading.Thread(target=self.close_output)
        output_thread.start()

    def close_output(self):
        """
        功能：释放相关资源
        """
        while self.__play_flag is False:
            self.__output_stream.stop_stream()
            self.__output_stream.close()
            self.__output_audio.terminate()
            time.sleep(0.1)
        self.__play_flag = True

    def play_wav_file(self):
        """
        功能:播放wav文件
        """
        self.play_file_thread()
        self.close_output_thread()

    def start_record_thread(self, path):
        """
        功能:开启录音线程
        参数1:需要报文文件的路径
        """
        self.__output_file = path
        record_thread = threading.Thread(target=self.start_sound_record)
        record_thread.start()

    def start_sound_record(self):
        """
        功能:开启录音函数
        """
        self.__frame = list()
        while self.__record_state:
            data = self.__input_stream.read(self.__chunk, exception_on_overflow=False)  # exception_on_overflow防止出现阻塞问题
            if data != b'':
                self.__frame.append(data)

    def stop_sound_record(self):
        """
        功能：停止录音
        """
        time.sleep(1)  # 暂停1秒,防止有未录入的数据
        self.__record_state = False
        self.close_stream()
        self.save_sound_record()

    def close_stream(self):
        """
        功能:关闭音频流
        """
        if self.__record_state is False:  # 判断是否已停止录制，此处如果直接执行，会存在录制线程还未停止的问题
            self.__input_stream.stop_stream()
            self.__input_stream.close()
            self.__input_audio.terminate()

    def save_sound_record(self):
        """
        功能:保存录音文件
        """
        wave_file = wave.open(self.__output_file, 'wb')
        wave_file.setnchannels(self.__channels)
        wave_file.setsampwidth(self.__input_audio.get_sample_size(self.__sample_format))
        wave_file.setframerate(self.__sample_rate)
        wave_file.writeframes(b''.join(self.__frame))
        wave_file.close()


if __name__ == '__main__':
    sound_record = SoundRecord()
    # audio_state = sound_record.open_audio_input_stream()  # 录制音频
    # if audio_state is True:
    #     sound_record.start_record_thread("testcode.wav")
    #     time.sleep(10)
    #     sound_record.stop_sound_record()
    # else:
    #     print("设备未连接")
    audio_output_state = sound_record.open_audio_output_stream(r'testcode.wav')  # 播放录音
    if audio_output_state is True:
        sound_record.play_wav_file()
    else:
        print("扬声器被占用")

写在最后

我是一名车载测试开发工程师，希望能和志同道合的朋友一起相互学习进步