I use pocketsphinx-0.8 in C++ code.
1. With the use of these models and the dictionary recognition rate decreases dramatically (audio file containing one word with a duration of 1 second is recognized 8-10 seconds).
Function ps_process_raw takes too long. Here is my code.
2. I get confidence score as described here
But sometimes the confidence score becomes more than one. Why it could happen?
Thanks in advance for help.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I use pocketsphinx-0.8 in C++ code.
1. With the use of these models and the dictionary recognition rate decreases dramatically (audio file containing one word with a duration of 1 second is recognized 8-10 seconds).
Function ps_process_raw takes too long.
Here is my code.
2. I get confidence score as described here
But sometimes the confidence score becomes more than one. Why it could happen?
Thanks in advance for help.
We recommend to use latest version 5 prealpha
You need to provide test data (audio) and decoder log in order to get help on this issue.