search

hjkwon0609 / source_separation_ml_jeju

Source Separation Project For ML Jeju Camp 2017
48 stars 17 forks source link

In run.py stack_spectrograms(spec) #2

Open peterlee909 opened 6 years ago

peterlee909 commented 6 years ago

According to the idea of paper ''' Stack spectrograms so that each element in the spectrogram now has 3 elements (prev_frame, curr_frame, next_frame) For the first(last) frame, prev_frame(next_frame) is zero vector ''' stacked = tf.stack([tf.pad(spec,[[0, 2], [0, 0]], "CONSTANT"), tf.pad(spec,[[1, 1], [0, 0]], "CONSTANT"), tf.pad(spec,[[0, 2], [0, 0]], "CONSTANT")], axis=2) # axis=0

I think the code should be: stacked = tf.stack([tf.pad(spec,[[2, 0], [0, 0]], "CONSTANT"), tf.pad(spec,[[1, 1], [0, 0]], "CONSTANT"), tf.pad(spec,[[0, 2], [0, 0]], "CONSTANT")], axis=2) # axis=0

Am I correct?

youngsuenXMLY commented 6 years ago

@peterlee909 I agree.

ucasiggcas commented 5 years ago

hi, Did you get success for the project ? and where is the dataset ?

thx

peterlee909 commented 5 years ago

@ucasiggcas No. But I implement by myself.

The dataset is here: https://sites.google.com/site/unvoicedsoundseparation/mir-1k