search

llSourcell / tensorflow_speech_recognition_demo

This is the code for 'How to Make a Simple Tensorflow Speech Recognizer' by @Sirajology on Youtube
384 stars 248 forks source link

How to realise the output prediction #33

Open raffaelbd opened 5 years ago

raffaelbd commented 5 years ago

After editing the #7 & #30 line, the code gives some output but in array format like [[0.08132637 0.09584257 0.0770199 0.0861607 0.10533866 0.12470371 0.1259584 0.10570324 0.08132752 0.11661901]. so from this how to realise the prediction of word & how someone know that this ans predicts which audio file.

raffaelbd commented 5 years ago

i edited only two lines. 1."training_iters = 300000 " is replaced by "training_iters = 10". 2."while 1: #trainingiters" is replaced by "for in range(10 ):" note that here ,training step will be 100,that means 10 times of 10. iteration & range no is also 10.so training step will be the 10 times of iteration no . i also removed some code to avoid the "value error at least two variables..." problem.