python3——音频处理（MAC）

1、将pcm音频文件转为mp3、wav格式，以便在web浏览器播放音频

def pcm2mp3(pcm_bytes, save_path, channels=1, bits=16, sample_rate=16000):
        """
        desc: 将pcm字节存储为mp3、wav文件
        """
        import wave
        import os
        from pydub import AudioSegment
        pcm_dir, pcm_name = os.path.split(save_path)
        pcm_dir = os.path.dirname(save_path)
        pcm_name = pcm_name.split('.')[0]
        wave_path = os.path.join(pcm_dir, pcm_name + '.wav')
        with wave.open(wave_path, 'wb') as wavfile:
            # 设置声道数
            wavfile.setnchannels(channels)
            # 设置采样位宽
            wavfile.setsampwidth(bits // 8)
            # 设置采样率
            wavfile.setframerate(sample_rate)
            # 写入 data 部分
            wavfile.writeframes(pcm_bytes)
         
        mp3_path = os.path.join(pcm_dir, pcm_name + '.mp3')
        wf = AudioSegment.from_wav(wave_path)
        wf.export(mp3_path, format="mp3")
    
def pcm2mp3_fromfile(pcm_file, channels=1, bits=16, sample_rate=16000):
        # https://www.modb.pro/db/533986
        # 打开 PCM 文件
        with open(pcm_file, 'rb') as f:
            pcm_data = f.read()
        # 打开将要写入的 WAVE 文件
        if pcm_data is None:
            return
        pcm2mp3(pcm_data, pcm_file, channels, bits, sample_rate)

pcm2mp3_fromfile("./report/out.pcm")

2、录制音频，生成pcm文件

import pyaudio
import wave
import os
import numpy as np

def record_audio(filename, duration, sample_rate=16000, channels=1, chunk=1024):
    audio_format = pyaudio.paInt16
    audio = pyaudio.PyAudio()
    stream = audio.open(format=audio_format,
                        channels=channels,
                        rate=sample_rate,
                        input=True,
                        frames_per_buffer=chunk)
    frames = []
    print("开始录制音频...")
    for i in range(0, int(sample_rate / chunk * duration)):
        data = stream.read(chunk)
        frames.append(data)
    print("录制完成！")
    stream.stop_stream()
    stream.close()
    audio.terminate()
    wf = wave.open(filename, 'wb')
    wf.setnchannels(channels)
    wf.setsampwidth(audio.get_sample_size(audio_format))
    wf.setframerate(sample_rate)
    wf.writeframes(b''.join(frames))
    wf.close()

def wav2pcm(filename):
    """
    desc: wav转为pcm
    """
    # import soundfile as sf
    pcm_file_path = filename.split('.')[0] + '.pcm'
    with open(filename, 'rb') as f:
        f.seek(0)
        f.read(44)
        data = np.fromfile(f, dtype=np.int16)
        data.tofile(pcm_file_path)

def play_audio(filename):
    """
    desc: 播放
    """
    audio = pyaudio.PyAudio()
    wf = wave.open(filename, 'rb')
    stream = audio.open(format=audio.get_format_from_width(wf.getsampwidth()),
                        channels=wf.getnchannels(),
                        rate=wf.getframerate(),
                        output=True)
    chunk = 1024
    data = wf.readframes(chunk)
    print("开始播放音频...")
    while data:
        stream.write(data)
        data = wf.readframes(chunk)
    print("播放完成！")
    stream.stop_stream()
    stream.close()
    audio.terminate()

record_audio('test.wav', 5)
play_audio('test.wav')
wav2pcm('test.wav')

3、基于文本合成音频

def tts():
    """
    desc: 语音合成
    """
    import pyttsx3
    engine = pyttsx3.init()
    engine.say('今天天气怎么样')
    engine.setProperty('rate', 125)
    engine.setProperty('volume', 1.0)
    engine.runAndWait()
    engine.stop()