The enhanced .wav file sounds very weird!

[ANYMS-A]

After run run_segan+_clean.sh I obtained enhanced .wav files, but it sounds like the audio is slowed down by many times. I'd like to know which test data set should I use?

[santi-pdp]

Hi there, please make sure you have audio data sampled at 16kHz in your noisy files.

[ANYMS-A]

Thanks! I use the noisy test set of noisy speech dataset of Edinburgh, So I thought a resampling is not needed. Do you mean, I have to resample the raw data first?

[santi-pdp]

You can check the sampling rate of your files with soxi command:

soxi data/LibriTTS/dev-clean/1272/128104/1272_128104_000001_000000.wav 

Input File     : 'data/LibriTTS/dev-clean/1272/128104/1272_128104_000001_000000.wav'
Channels       : 1
Sample Rate    : 24000
Precision      : 16-bit
Duration       : 00:00:01.52 = 36480 samples ~ 114 CDDA sectors
File Size      : 73.0k
Bit Rate       : 384k
Sample Encoding: 16-bit Signed Integer PCM

[ANYMS-A]

Many thanks! I followed your instruction to resample the audio in 16kHz, then I can't hear anything from the enhanced speech, all files are play to end in silence. The resampled audio heard fine. Should I change any other parameters?

[LucySha]

make sure the input audio with the format:16KHz, and 16bit, if the noisy audio with 16Khz and 32bit, you couldn't hear anything from the enhanced speech.

[unanan]

After run run_segan+_clean.sh I obtained enhanced .wav files, but it sounds like the audio is slowed down by many times. I'd like to know which test data set should I use?

#22 (comment)

[jellyfish1456]

very weird