How can I save the reduced noise in a wav file?? about noisereduce

[timsainb]

You are probably saving in the wrong format. E.g. saving data in int16 as float32 or vice versa. Try loading the saved data back up in python using scipy.io and take a look at the data.

[abhinashsivan]

load data

rate, data = scipy.io.wavfile.read("noisy_audio.wav")

selecting whole file to perform the noise reduction

noisy_part = data

perform noise reduction

reduced_noise = nr.reduce_noise(audio_clip=data, noise_clip=noisy_part, verbose=False)

save the noise removed data as wav

scipy.io.wavfile.write('clean_audio.wav', rate, reduced_noise)

[asarsembayev]

scipy.io.wavfile.write('clean_audio.wav', rate, reduced_noise)

I have the same issue as @jcabrerahi . I tried scipy.io.write(), librosa.output.write_wav(), wavfile.write(). Nothing changes only plain noise. However, I obtained cleaned and good audio and I can listen to it in IPython.display.Audio(data=reduced_noise, rate=rate)
Anyone knows what's the problem?

[asarsembayev]

You are probably saving in the wrong format. E.g. saving data in int16 as float32 or vice versa. Try loading the saved data back up in python using scipy.io and take a look at the data.

Indeed. Thanks a lot @timsainb !
scipy.io.wavfile.write('1.wav', rate, np.asarray(reduced_noise, dtype=np.int16)) worked for me.
@abhinashsivan , try to follow too.

[abhinashsivan]

Sure, Thanks.