Skip to content

Filtering speech sentences using multiple MOS scoring tools

Notifications You must be signed in to change notification settings

nanless/speech-MOS-filter

Repository files navigation

speech-MOS-filter

Filtering speech sentences using multiple MOS scoring tools

Usage

python compute_mos.py --wav_files_dir /path/to/wavfile_directory --output_statistics_dir /path/to/output_statistics_directory
python hisgram.py --statistics_folder_path /path/to/output_statistics_directory
python filter.py --statistics_folder_path /path/to/output_statistics_directory --origianl_folder_path /path/to/original_wavfile_directory --filtered_folder_path /path/to/filtered_wavfile_directory

You can set the threshold values for each MOS scoring tool in the filter.py file.

Credits

Special thanks to the following projects:

DNSMOS

NISQA

SIGMOS

UTMOS

WVMOS

About

Filtering speech sentences using multiple MOS scoring tools

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages