search

microsoft / MS-SNSD

The Microsoft Scalable Noisy Speech Dataset (MS-SNSD) is a noisy speech dataset that can scale to arbitrary sizes depending on the number of speakers, noise types, and Speech to Noise Ratio (SNR) levels desired.
MIT License
468 stars 142 forks source link

some bugs in audiolib #8

Closed kangcp closed 4 years ago

kangcp commented 4 years ago

I think in audiolib file, line 66 code: noisescalar = np.sqrt (rmsclean / (10 ** (snr / 20)) / rmsnoise), snr should be divided by 10, not 20.

Liu-1994 commented 4 years ago

I also think there is something wrong. But maybe it should be "noisescalar = rmsclean / (10 ** (snr / 20)) / rmsnoise", as the "rmsclean" and "rmsnoise" have been square root.

chandanka90 commented 4 years ago

Recomputed rmsclean and rmsnoise after scaling. Should be good now.