java实现spectrogram函数_梅尔频谱(mel-spectrogram)提取，griffin_lim声码器【python代码分析】...

在语音分析，合成，转换中，第一步往往是提取语音特征参数。

利用机器学习方法进行上述语音任务，常用到梅尔频谱。

本文介绍从音频文件提取梅尔频谱，和从梅尔频谱变成音频波形。

从音频波形提取Mel频谱：

对音频信号预加重、分帧和加窗

对每帧信号进行短时傅立叶变换STFT，得到短时幅度谱

短时幅度谱通过Mel滤波器组得到Mel频谱

从Mel频谱重建音频波形

Mel频谱转换成幅度谱

griffin_lim声码器算法重建波形

去加重

声码器有很多种，比如world,straight等，但是griffin_lim是特殊的，它不需要相位信息就可以重频谱重建波形，实际上它根据帧之间的关系估计相位信息。和成的音频质量也较高，代码也比较简单。

音频波形 到 mel-spectrogram

sr = 24000 # 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 = 512 # 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 = 15

1

2

3

4

5

6

7

8

9

10

11

12

13

def 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)

mag: A 2d array of shape (T, 1+n_fft/2)

'''

# Loading sound file

y, sr = librosa.load(fpath, sr=sr)

# Trimming

y, _ = librosa.effects.trim(y, top_db=top_db)

# Preemphasis

y = np.append(y[0], y[1:] - preemphasis * y[:-1])

# stft

linear = librosa.stft(y=y,

n_fft=n_fft,

hop_length=hop_length,

win_length=win_length)

# magnitude spectrogram

mag = np.abs(linear) # (1+n_fft//2, T)

# mel spectrogram

mel_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 decibel

mel = 20 * np.log10(np.maximum(1e-5, mel))

mag = 20 * np.log10(np.maximum(1e-5, mag))

# normalize

mel = 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)

# Transpose

mel = mel.T.astype(np.float32) # (T, n_mels)

mag = mag.T.astype(np.float32) # (T, 1+n_fft//2)

return mel, mag

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

mel-spectrogram 到 音频波形

def melspectrogram2wav(mel):

'''# Generate wave file from spectrogram'''

# transpose

mel = mel.T

# de-noramlize

mel = (np.clip(mel, 0, 1) * max_db) - max_db + ref_db

# to amplitude

mel = np.power(10.0, mel * 0.05)

m = _mel_to_linear_matrix(sr, n_fft, n_mels)

mag = np.dot(m, mel)

# wav reconstruction

wav = griffin_lim(mag)

# de-preemphasis

wav = signal.lfilter([1], [1, -preemphasis], wav)

# trim

wav, _ = librosa.effects.trim(wav)

return wav.astype(np.float32)

def spectrogram2wav(mag):

'''# Generate wave file from spectrogram'''

# transpose

mag = mag.T

# de-noramlize

mag = (np.clip(mag, 0, 1) * max_db) - max_db + ref_db

# to amplitude

mag = np.power(10.0, mag * 0.05)

# wav reconstruction

wav = griffin_lim(mag)

# de-preemphasis

wav = signal.lfilter([1], [1, -preemphasis], wav)

# trim

wav, _ = librosa.effects.trim(wav)

return wav.astype(np.float32)

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

几个辅助函数：

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 * phase

X_t = invert_spectrogram(X_best)

y = np.real(X_t)

return y

def invert_spectrogram(spectrogram):

'''

spectrogram: [f, t]

'''

return librosa.istft(spectrogram, hop_length, win_length=win_length, window="hann")

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

预加重：

语音信号的平均功率谱受声门激励和口鼻辐射影响，高频端约在800HZ以上按6dB/倍频程衰落，预加重的目的是提升高频成分，使信号频谱平坦化，以便于频谱分析或声道参数分析．

---------------------