Motaz Saad - 2018-09-05

Hello,

I am trying to use US English models at https://sourceforge.net/projects/cmusphinx/files/Acoustic%20and%20Language%20Models/US%20English/ with pocketsphinx (python package) to decode English wav files, but I got the error "ERROR: "dict.c", line 195: Line 283: Phone ....... is mising in the acoustic model; word ..... ignored"
for so many phones. Please see a example of the error below:

ERROR: "dict.c", line 195: Line 283: Phone 'AH0' is mising in the acoustic model; word 'abort' ignored

I tried cmusphinx-en-us-5.2 and cmusphinx-en-us-5.1 and I had the same issue.
The decoder output is hm hmm ....... for all waves!

I use the dictionary from https://github.com/cmusphinx/cmudict
and the language model is from I am trying to use US English models at https://sourceforge.net/projects/cmusphinx/files/Acoustic%20and%20Language%20Models/US%20English/

Is guess it is AM problem rather than dictionary problem, please correct me if I am wrong.

I would appreciate your help
Best regards,
Motaz