N-HANS

icon company EIHW, University Augsburg icon location Augsburg, Germany icon location is born for in-the-wild audio enhancement, including speech, music, and general audio denoising, separation, and selective noise or source suppression.

Results 3 comments of N-HANS

Reverb + Higher Sample Rates

The existing N-HANS models were trained with 16kHz as an example. To have a model for other sampling rates, please modify two stuff in "reader.py" and "main.py". 1. sampling frequency:...

Reverb + Higher Sample Rates

If I understand your question correctly, for reverb speech you can have Noise = Reverb_Speech - Clean_Speech, this may require some preprocessing, and fixed the SNR=0dB in reader.py Speaking of...

Colab

Hello, currently we haven’t considered colab notebook, but it is a nice suggestion. ; )