The timestamps lag behind the actual audio by about 0.4 seconds. Although its a small offset, it's critical to the project that the start/end are very precise. Is there any way to increase the accuracy of pocketsphinx_continuous, or is there a more precise option out there?
Thanks
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I need to extract the timestamps of the beginnings and endings of words in an audio file. However, when I run
The timestamps lag behind the actual audio by about 0.4 seconds. Although its a small offset, it's critical to the project that the start/end are very precise. Is there any way to increase the accuracy of pocketsphinx_continuous, or is there a more precise option out there?
Thanks
Please provide the sample file and exact word which you think is not correctly detected.
Please provide the information about pocketsphinx version you are using.