I got this MFCC error when I running Deep Speech
Missing 5 bands starting at 0 in mel-frequency design. Perhaps too many channels or not enough frequency resolution in spectrum. (input_length: 257 input_sample_rate: 44100 output_channel_count: 40 lower_frequency_limit: 20 upper_frequency_limit: 4000
How I should do to improve frequency?
I have 2 seconds in each wav file.
Thank you
lissyx
((slow to reply) [NOT PROVIDING SUPPORT])
2
It is at training from Epoch 0. I am using transfer learning 2 branch. My utterances has English and Chinese mixed, but English occupied more than Chinese words. @lissyx
lissyx
((slow to reply) [NOT PROVIDING SUPPORT])
4
Is your whole dataset using 44.1kHz stereo ? have you made any change to the code ?
So the whole dataset has to be 16khz instead of 44.1kHz? @lissyx
lissyx
((slow to reply) [NOT PROVIDING SUPPORT])
8
No, the whole dataset has to be in the same format. Our default setting and what we train on is 16kHz, so it’s easier for you. Just don’t mix different sample rates.