m0_43395719的博客

读取音频import librosaimport pyworldsound, _ = librosa.load(wav_path, sr=16000)print(f'sound.shape = {sound.shape}') #sound.shape = (80000,)提取基频F0sr = 16000#第一种_f0, t = pw.dio(sound, sr) # raw pitch extractorf0 = pw.stonemask(sound, _f0, t, s.

1.导入语音import librosaaudio_path = './1.wav'x , sr = librosa.load(audio_path)print(f'x = {x}')print(f'x.shape = {x.shape}') # （采样率*语音时间，）print(f'sr = {sr}')# 采样率2. 修改采样率y , sr = librosa.load(audio_path， sr=new_sr)3. stft 短时傅里叶变换源码libroas.stft(y,

读取/导出 语音from pydub import AudioSegmentimport mathfrom pydub.generators import WhiteNoise#读取音频orgin_wav = AudioSegment.from_wav(path_file)#导出音频target_wav.export("D:/data/combin1.wav", format="wav")获取音频信息# 获取音频持续时间 单位 Sduration = orgin_wav.d.