Files used for bachelor's project titled "Persian Keyword Recognition by applying appropriate phonetic mapping and modifications in the Allosaurus decoder" (English title)
The Allosaurus folder contains the allosaurus model codes. Other files are the extra files used for processing the audio files and evaluating the models. Unfortunately, I cannot upload the audio files used for training the models since they have a license.
The final trained models in this project can be found at https://drive.google.com/drive/folders/1jMFa-eJ9yIIlKyFkUfABIi6b8aLZKGMN In order to use the models, they need to be unziped and copied to the path .\allosaurus\pretrained (relative to this repo)