def prediction_mode(in_dir, file_name, out_dir):
#Example: python main.py --mode="prediction"
#path to find pre-trained weights / save models
weights_path = './weights'
#pre trained model
name_model = 'model_unet'
#directory where read noisy sound to denoise
audio_dir_prediction = in_dir
#directory to save the denoise sound
dir_save_prediction = out_dir
#Name noisy sound file to denoise
audio_input_prediction = [file_name]
#Name of denoised sound file to save
audio_output_prediction = file_name
# Sample rate to read audio
sample_rate = 8000
# Minimum duration of audio files to consider
min_duration = 1.0
#Frame length for training data
frame_length = 8064
# hop length for sound files
hop_length_frame = 8064
#nb of points for fft(for spectrogram computation)
n_fft = 255
#hop length for fft
hop_length_fft = 63
prediction(weights_path, name_model, audio_dir_prediction,
dir_save_prediction, audio_input_prediction,
audio_output_prediction, sample_rate, min_duration,
frame_length, hop_length_frame, n_fft, hop_length_fft)
elif prediction_mode:
#Example: python main.py --mode="prediction"
#path to find pre-trained weights / save models
weights_path = args.weights_folder
#pre trained model
name_model = args.name_model
#directory where read noisy sound to denoise
audio_dir_prediction = args.audio_dir_prediction
#directory to save the denoise sound
dir_save_prediction = args.dir_save_prediction
#Name noisy sound file to denoise
audio_input_prediction = args.audio_input_prediction
#Name of denoised sound file to save
audio_output_prediction = args.audio_output_prediction
# Sample rate to read audio
sample_rate = args.sample_rate
# Minimum duration of audio files to consider
min_duration = args.min_duration
#Frame length for training data
frame_length = args.frame_length
# hop length for sound files
hop_length_frame = args.hop_length_frame
#nb of points for fft(for spectrogram computation)
n_fft = args.n_fft
#hop length for fft
hop_length_fft = args.hop_length_fft
prediction(weights_path, name_model, audio_dir_prediction, dir_save_prediction, audio_input_prediction,
audio_output_prediction, sample_rate, min_duration, frame_length, hop_length_frame, n_fft, hop_length_fft)
# create_data(noise_dir, voice_dir, dataset_noise, dataset_voice, time_wave_dir, sound_dir,
# spectrogram_dir, sample_rate, min_duration, frame_length, hop_length_frame,
# hop_length_frame_noise, nb_samples, n_fft, hop_length_fft, list_noise_files,
# list_voice_files)
training(dataset_noise, dataset_voice, spectrogram_dir,
weights_dir, model_name, training_from_scratch, epochs,
batch_size)
elif prediction_mode:
weights_dir = args.weights_dir
model_name = args.model_name
input_dir = args.input_dir
output_dir = args.output_dir
sample_rate = args.sample_rate_predict
min_duration = args.min_duration
frame_length = args.frame_length
hop_length_frame = args.hop_length_frame
n_fft = args.n_fft
hop_length_fft = args.hop_length_fft
prediction(weights_dir, model_name, input_dir, output_dir, sample_rate,
frame_length, hop_length_frame, n_fft, hop_length_fft)
elif plot_mode:
temp_dir = args.temp_dir
sample_rate = args.sample_rate_plot
hop_length_fft = args.hop_length_fft
plot_out(temp_dir, sample_rate, hop_length_fft)