梅尔频谱图与音频相互转化

看大佬写的

import scipy.signal as signal
import librosa
import torch
import numpy as np
import copysr = 22050 # Sample rate.
n_fft = 2048 # fft points (samples)
frame_shift = 0.0125 # seconds
frame_length = 0.05 # seconds
hop_length = int(sr*frame_shift) # samples.
win_length = int(sr*frame_length) # samples.
n_mels = 80 # Number of Mel banks to generate
power = 1.2 # Exponent for amplifying the predicted magnitude
n_iter = 100 # Number of inversion iterations
preemphasis = .97 # or None
max_db = 100
ref_db = 20
top_db = 15def get_spectrograms(fpath):'''Returns normalized log(melspectrogram) and log(magnitude) from `sound_file`.Args:sound_file: A string. The full path of a sound file.Returns:mel: A 2d array of shape (T, n_mels) <- Transposedmag: A 2d array of shape (T, 1+n_fft/2) <- Transposed'''# Loading sound filey, sr = librosa.load(fpath, sr=22050)# Trimmingy, _ = librosa.effects.trim(y, top_db=top_db)# Preemphasisy = np.append(y[0], y[1:] - preemphasis * y[:-1])# stftlinear = librosa.stft(y=y,n_fft=n_fft,hop_length=hop_length,win_length=win_length)# magnitude spectrogrammag = np.abs(linear)  # (1+n_fft//2, T)# mel spectrogrammel_basis = librosa.filters.mel(sr, n_fft, n_mels)  # (n_mels, 1+n_fft//2)mel = np.dot(mel_basis, mag)  # (n_mels, t)# to decibelmel = 20 * np.log10(np.maximum(1e-5, mel))mag = 20 * np.log10(np.maximum(1e-5, mag))# normalizemel = np.clip((mel - ref_db + max_db) / max_db, 1e-8, 1)mag = np.clip((mag - ref_db + max_db) / max_db, 1e-8, 1)# Transposemel = mel.T.astype(np.float32)  # (T, n_mels)mag = mag.T.astype(np.float32)  # (T, 1+n_fft//2)return mel, magdef melspectrogram2wav(mel):'''# Generate wave file from spectrogram'''# transposemel = mel.T# de-noramlizemel = (np.clip(mel, 0, 1) * max_db) - max_db + ref_db# to amplitudemel = np.power(10.0, mel * 0.05)m = _mel_to_linear_matrix(sr, n_fft, n_mels)mag = np.dot(m, mel)# wav reconstructionwav = griffin_lim(mag)# de-preemphasiswav = signal.lfilter([1], [1, -preemphasis], wav)# trimwav, _ = librosa.effects.trim(wav)return wav.astype(np.float32)def spectrogram2wav(mag):'''# Generate wave file from spectrogram'''# transposemag = mag.T# de-noramlizemag = (np.clip(mag, 0, 1) * max_db) - max_db + ref_db# to amplitudemag = np.power(10.0, mag * 0.05)# wav reconstructionwav = griffin_lim(mag)# de-preemphasiswav = signal.lfilter([1], [1, -preemphasis], wav)# cwav, _ = librosa.effects.trim(wav)return wav.astype(np.float32)def _mel_to_linear_matrix(sr, n_fft, n_mels):m = librosa.filters.mel(sr, n_fft, n_mels)m_t = np.transpose(m)p = np.matmul(m, m_t)d = [1.0 / x if np.abs(x) > 1.0e-8 else x for x in np.sum(p, axis=0)]return np.matmul(m_t, np.diag(d))def griffin_lim(spectrogram):'''Applies Griffin-Lim's raw.'''X_best = copy.deepcopy(spectrogram)for i in range(n_iter):X_t = invert_spectrogram(X_best)est = librosa.stft(X_t, n_fft, hop_length, win_length=win_length)phase = est / np.maximum(1e-8, np.abs(est))X_best = spectrogram * phaseX_t = invert_spectrogram(X_best)y = np.real(X_t)return ydef invert_spectrogram(spectrogram):'''spectrogram: [f, t]'''return librosa.istft(spectrogram, hop_length, win_length=win_length, window="hann")def plot_spectrogram_to_numpy(spectrogram):fig, ax = plt.subplots(figsize=(12, 3))im = ax.imshow(spectrogram, aspect="auto", origin="lower",interpolation='none')plt.colorbar(im, ax=ax)plt.xlabel("Frames")plt.ylabel("Channels")plt.tight_layout()fig.canvas.draw()data = save_figure_to_numpy(fig)plt.close()return datadef save_figure_to_numpy(fig):# save it to a numpy array.data = np.fromstring(fig.canvas.tostring_rgb(), dtype=np.uint8, sep='')data = data.reshape(fig.canvas.get_width_height()[::-1] + (3,))return dataif __name__ == '__main__':# melpost = torch.load('E:\\tacotron2\\melpost.wav(1).pt',map_location='cpu').detach().numpy()# mel = np.zeros((80, 1025))# mel[:, :960] = melpost# mel[:, 960:] = melpost[:,:65]aa = get_spectrograms('E:\\tacotron2\\000005.wav')import matplotlib.pyplot as plt# mel = mel / 11 + 1plt.figure()# plt.subplot(3, 1, 1)# plt.imshow(plot_spectrogram_to_numpy(mel))plt.subplot(2, 1, 2)plt.imshow(plot_spectrogram_to_numpy(aa[1].T))plt.subplot(2, 1, 1)plt.imshow(plot_spectrogram_to_numpy(aa[0].T))plt.show()# wav = melspectrogram2wav(mel.T)wav1 = melspectrogram2wav(aa[0])# librosa.output.write_wav("gg_stft.wav", wav, sr)librosa.output.write_wav("gg_stf11.wav", wav1, sr)

梅尔频谱图与音频相互转化相关推荐

python绘制频谱图,在音频分析中绘制频谱图
I am working on speech recognition using neural network. To do so I need to get the spectrograms of ...
数字信号处理实验三用fft对信号作频谱分析_机器学习中的音频特征：理解Mel频谱图...
如果你像我一样,试着理解mel的光谱图并不是一件容易的事.你读了一篇文章,却被引出了另一篇,又一篇,又一篇,没完没了.我希望这篇简短的文章能澄清一些困惑,并从头解释mel的光谱图. 信号信号是一定量 ...
基于梅尔频谱的音频信号分类识别(Pytorch)
本项目将使用Pytorch,实现一个简单的的音频信号分类器,可应用于机械信号分类识别,鸟叫声信号识别等应用场景. 项目使用librosa进行音频信号处理,backbone使用mobilenet_v2, ...
如何使用卷积神经网络从梅尔谱图检测 COVID-19 咳嗽
新冠肺炎 COVID-19(2019 年冠状病毒病)是一种会导致人类呼吸系统问题.体温超过 38°C 的发烧.呼吸急促和咳嗽的疾病.即使是这种疾病也会导致肺炎死亡. 在 COVID-19 之前被认为是 ...
librosa | 梅尔谱图最幽默的解释
文章目录写在前面正文开始了解梅尔频谱图几天前发生在我脑海中的真实对话频谱图梅尔量表梅尔频谱图回顾未完待续... 小白进来! 写在前面 Medium的一篇文章,特别幽默地介绍了梅尔谱图 ...
深入理解梅尔刻度、梅尔滤波器组和梅尔时频谱图
前情提要短时傅里叶变换公式 S ( m , k ) = ∑ n = 1 N − 1 x ( n + m H ) w ( n ) e − i 2 π k N n S(m,k) = \sum_{n=1} ...
pyhton 画出音频文件的波形图和频谱图
pyhton 画出音频文件的波形图和频谱图 # -*- coding:utf-8 -*- import wave import struct import numpy as np import mat ...
音频处理相关内容学习——自动编码器——变分自动编码器——频谱图
文章目录概述一.Approaches And Challenges 生成的声音是什么类型训练模型使用的是什么特征原始音频Raw Audio 频谱图Spectrograms 声音生成模型常用的结 ...
音频频谱图（自定义View——进阶篇2）
音频频谱 + 波形图绘制我们经常有看到音乐播放器播放界面会有频谱图显示,感觉很炫,今天我就带大家来实现频谱图,顺便将波形图绘制也分享给大家,这里重点讲频谱这块.我们这里的频谱采用8分频fft,这里 ...

梅尔频谱图与音频相互转化

梅尔频谱图与音频相互转化相关推荐

最新文章

热门文章