import soundfile as sf
import numpy as np
from scipy import signal as scisig
from scipy import fftpack as scifft
from crying_vegetable import vocoder, reduct_frac
from crying_vegetable.mattoolbox import signal as matsig
from crying_vegetable import vegetablecry as vc
from IPython.display import Audio
from matplotlib import pyplot as plt
import japanize_matplotlib
#import SciencePlots
plt.style.use(['science', 'ieee'])
#plt.style.use(['science', 'cjk-jp-font', 'no-latex'])
from matplotlib import rc
#rc('text', usetex=True)

from os import path

jikken_tsv = 'data/jikken2-a.txt'
time_prefix = 'jikken8.200'
#jikken_tsv = 'data/jikken3.txt'
#time_prefix = 'jikken300'

time_unit = 'minutes'

# duration_change_rate
fs_data = 24       # in [1/s] 

# 出力する音響信号について
fs_out = 44100      # in [1/s]

#type_out = 'chirp'
#type_out = 'male_a'
#type_out = 'male_i'
#type_out = 'male_u'
#type_out = 'male_e'
#type_out = 'male_o'
#type_out = 'female_a'
#type_out = 'female_i'
#type_out = 'female_u'
type_out = 'female_e'
#type_out = 'female_o'

jikken_prefix = path.splitext(path.basename(jikken_tsv))[0]

t, f = vc.read_jikken(jikken_tsv,time_prefix,time_unit) 
tr, fr = vc.resample_jikken(t,f) # 欠損補間

fs_data0 = 1/(tr[1]-tr[0])

fig = plt.figure()
ax1 = fig.add_subplot(1,1,1)
ax1.plot(tr,fr)
ax1.set_ylim([-1,1])
ax1.set_xlabel('Time [sec.]')
ax1.set_ylabel('NDI [?]')
ax1.set_title(jikken_tsv)
ax1.grid()
TimeVariantNDI_png = 'fig/' + jikken_prefix + '_fig_TimeVariantNDI.png'
fig.savefig(TimeVariantNDI_png)

if type_out == 'chirp':
    gender = 'chirp'
    vowel = 'broad'
    target_freq_limits = [200,3500]
else:
    gender, vowel = type_out.split('_')
    if gender == 'female':
        target_freq_limits = [168, 880]
    elif gender == 'male':
        target_freq_limits = [84, 440]

jikken_wav = 'wav/' + jikken_prefix + '_' + gender + '_' + vowel + '.wav'

print(gender,vowel,target_freq_limits)

f2 = matsig.scaling(fr, [-1,+1], target_freq_limits)

(p,q) = reduct_frac.reduct_frac(fs_out,fs_data)
f2ud = matsig.upsampling(f2,p,q)

t2 = np.arange(len(f2ud))/fs_out

fig = plt.figure()
ax = fig.add_subplot(111)
ax.plot(t2,f2ud)
ax.grid()
ax.set_ylim(target_freq_limits)
ax.set_xlabel('Time [sec.]')
ax.set_ylabel('Frequency [Hz]')
ax.set_title(f'Time variant frequencies of Vocoded Vegetable Cry\n{gender}-/{vowel}/')
TimeVariantFreq_png = 'fig/' + jikken_prefix + f'_fig_TimeVariantFreq_{gender}_{vowel}.png'
fig.savefig(TimeVariantFreq_png)

instant_phase = np.cumsum(f2ud)/fs_out
y = np.sin(2*np.pi*instant_phase)

if type_out != 'chirp':
    vocal = np.full((len(t2),),0.0)
    vocal[np.where(y>0.95)[0]]=1.0
    y = vocoder.formant_filter(vocal,fs_out,gender,vowel)

time_region = [2.5, 2.6]
point_region = np.array(time_region)*fs_out
point_region = point_region.astype('int')
print(f'{point_region}')
V = scifft.fft(y[point_region[0]:point_region[1]])
lx = len(V)
Mag = 10*np.log10(np.real(V*np.conj(V)))
freq = np.arange(lx)/lx*fs_out
nlx = lx//2

fig = plt.figure()
ax = fig.add_subplot(111)
ax.loglog(freq[:nlx],Mag[:nlx])
#ax.set_xscale('log')
#ax.set_yscale('log')
ax.set_ylim([10**0, 10**2])
ax.set_xlabel('Frequency [Hz]')
ax.set_ylabel('Power [dB]')
ax.set_title(f'Power Spectrum {time_region[0]}--{time_region[1]}[sec.]')
ax.grid()

y /= np.amax(np.abs(y))
y *= 0.8 
sf.write(jikken_wav,y,fs_out)
Audio(jikken_wav, rate=fs_out)