I have a set of audio files for different pronounciation for a set of sentences. each sentence has several ways of pronouncing it .. The user will choose a sentence and say it but he have to follow the way of pronounciation of one of the audio files for the sentence. and the system should check if he said it correctly or not ...
so I think I don't need a dictionary or a language model but I want to train and the model by only the audio files (I think it will need only acoustic model)
Can I do this using sphinx? I mean can it work without having a dictionary or a language model?
Last edit: Rawan Laz 2017-02-01
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have a set of audio files for different pronounciation for a set of sentences. each sentence has several ways of pronouncing it .. The user will choose a sentence and say it but he have to follow the way of pronounciation of one of the audio files for the sentence. and the system should check if he said it correctly or not ...
so I think I don't need a dictionary or a language model but I want to train and the model by only the audio files (I think it will need only acoustic model)
Can I do this using sphinx? I mean can it work without having a dictionary or a language model?
Last edit: Rawan Laz 2017-02-01
http://cmusphinx.sourceforge.net/wiki/faq/#qhow_to_implement_pronunciation_evaluation