Hello, I already have a decoder that takes in features and outputs phonemes. Is there a way that I can skip the feature extraction and decoding with the acoustic model part and go straight to searching for word matches given phonemes?
The decoder I already have is in another language (it's not done in CMUSphinx)
Thanks :)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello, I already have a decoder that takes in features and outputs phonemes. Is there a way that I can skip the feature extraction and decoding with the acoustic model part and go straight to searching for word matches given phonemes?
No, you have to implement it yourself. CMUSphinx searches for words in audio, not for phonemes.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Okay. Say I implement everything in CMUSphinx. Will the decoder I already have written that uses a neural network take part in the process of recognition?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I am sorry, cmusphinx decoder works a bit different from the way you imagine. I doubt you will be able to integrate your phonetic decoder in cmusphinx.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello, I already have a decoder that takes in features and outputs phonemes. Is there a way that I can skip the feature extraction and decoding with the acoustic model part and go straight to searching for word matches given phonemes?
The decoder I already have is in another language (it's not done in CMUSphinx)
Thanks :)
No, you have to implement it yourself. CMUSphinx searches for words in audio, not for phonemes.
Okay. Say I implement everything in CMUSphinx. Will the decoder I already have written that uses a neural network take part in the process of recognition?
I am sorry, cmusphinx decoder works a bit different from the way you imagine. I doubt you will be able to integrate your phonetic decoder in cmusphinx.