Dionyssos's picture
format
7fa53df
raw
history blame
448 Bytes
from audiocraft.audiogen import AudioGen #, audio_write
import audiofile
import numpy as np
print('\n\n\n\n___________________')
txt = 'car'
sound_generator = AudioGen.get_pretrained('facebook/audiogen-medium')
sound_generator.set_generation_params(duration=1) # why is generating so long at 14 seconds
x = sound_generator.generate([txt])[0].detach().cpu().numpy()[0, :]
x /= np.abs(x).max() + 1e-7
audiofile.write('_audio1_.wav', x, 16000)