Use 11025 Hz model - Githubissues

maxrmorrison / pyfoal

Python forced alignment

MIT License

71 stars 4 forks source link

Closed maxrmorrison closed 3 years ago

maxrmorrison commented 3 years ago

This model achieves the highest accuracy according to the original paper "Speaker Identification on the SCOTUS corpus".

maxrmorrison commented 3 years ago

Fixed in b738f16