Synthetic chirp generator.

In [1]:

    

%matplotlib inline
import matplotlib.pyplot as plt
plt.rcParams['figure.figsize'] = (15, 4)


    

In [2]:

    

import numpy as np
import scipy
import scipy.signal
import librosa


    

In [3]:

    

# Settings.
sampling_freq = 384000
start_freq = 120000
end_freq = 40000
duration= 0.008
time_between_chirps = 0.1
max_amplitude = 0.3
noise_level = 0.002
number_of_chirps = 10
file_name = 'chirp_120_to_40_khz_TE384.wav'


    

In [4]:

    

# Create chirp.
time = np.linspace(0, duration, int(sampling_freq * duration))
chirp = scipy.signal.waveforms.chirp(time, 
                                     f0=start_freq, 
                                     f1=end_freq, 
                                     t1=duration, 
                                     method='quadratic', 
                                     vertex_zero=False)
len(chirp)


    

    
Out[4]:





3072

In [5]:

    

# Apply window function.
chirp = chirp * scipy.signal.hanning(len(time)) * max_amplitude


    

    




/home/parallels/Desktop/dev/w_cloudedbats/cloudedbats_dsp/venv/lib/python3.6/site-packages/ipykernel_launcher.py:2: DeprecationWarning: `hanning` is deprecated, use `scipy.signal.windows.hann` instead!
  

In [6]:

    

# Create silent part.
silent_duration = time_between_chirps - duration
silent = np.zeros(int(sampling_freq * silent_duration))
print(silent_duration)


    

    




0.092

In [7]:

    

# Build sequence.
signal = silent[:int(len(silent)/2)] # Start with half lenght silence.
for index in range(number_of_chirps):
    signal = np.concatenate((signal, chirp, silent))


    

In [8]:

    

# Add noise.
signal = signal + np.random.randn(len(signal)) * noise_level


    

In [9]:

    

# Plot signal.
plt.plot(signal)
plt.show()


    

In [10]:

    

# Write to file in Time Expanded mode.
freq = int(sampling_freq / 10)
librosa.output.write_wav(file_name, signal, freq)


    

In [ ]: