Closed hrishikeshathalye closed 2 years ago
Hmm, interesting; thank you for the head's up! I guess the '1' value comes from the code (which has '1' as the default also on that web page). Feel free to open a pull request to fix this!
Another small change. Also changed equation (7-27) considering the value of sT as 1.
In the mos_score function in util.py the value of Bpl used in the calculation is 1. Whereas, in the G.107 recommendation, the default recommended value is 4.3 and the permitted range is from 4.3 to 40. MOS values for some combinations of delay and loss are calculated incorrectly due to this. Setting Bpl to 4.3 will make the code conform exactly to the reference implementation given here https://www.itu.int/ITU-T/studygroups/com12/emodel/calculnb.php