Could you tell me if the MOS rating is objective or subjective?

[magicmans]

Could you please tell me if the MOS data in the NISQA Corpus is based on subjective MOS values from multi-person evaluations or on objective MOS values calculated using the ESPQ algorithm?

[gabrielmittag]

The MOS scores in the NISQA Corpus are subjective MOS values. Please also see the Wiki for more info, such as the number of ratings per file: https://github.com/gabrielmittag/NISQA/wiki/NISQA-Corpus

[magicmans]

Thank you for your response. Another question I have is: Can the NISQA model be trained and used for objective MOS prediction?

[gabrielmittag]

Yes, that's exactly what it is trained for. Please check the Readme for more information

[magicmans]

thanks